TECHNOLOGIES FOR DATA DEDUPLICATION IN DISAGGREGATED ARCHITECTURES

Abstract

Technologies for providing data deduplication in a disaggregated architecture include a network device. The network device is to receive, from a compute sled, a request to write a data block to one or more data storage sleds and determine, for each of one or more data sub-blocks within the data block and from deduplication data indicative of physical addresses of data sub-blocks, whether each data sub-block is already stored in a data storage device of a data storage sled. Additionally, the network device is to write, in the deduplication data and in response to a determination that a data sub-block is already stored in a data storage device, a pointer to a physical address of the already-stored data sub-block in association with a logical address of the data sub-block.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

The present application claims the benefit of U.S. Provisional Patent Application No. 62/427,268, filed Nov. 29, 2016 and Indian Provisional Patent Application No. 201741030632, filed Aug. 30, 2017.

BACKGROUND

Database volumes continue to grow at an unprecedented rate, significantly outpacing growth in data storage capacities of data storage devices. Some databases experience massive spikes in their sizes as they are being used (e.g., storing data, selecting data, reading data, comparing data, etc.) and the data storage capacities of the compute devices maintaining the databases must be able to absorb these spikes in order to operate without encountering errors, such as an inability to complete a requested operation on the data.

Typical data compaction techniques, such as deduplicating data, are designed to operate on the data storage system of a single compute device. As such, typical memory compaction techniques are not available to be applied to databases that are spread across multiple compute devices, such as in a cloud data center in which multiple disaggregated resources may be dynamically provisioned to execute workloads (e.g., applications, processes, etc.) that operate on databases that may be distributed across multiple data storage devices throughout a fabric (e.g., a network).

BRIEF DESCRIPTION OF THE DRAWINGS

The concepts described herein are illustrated by way of example and not by way of limitation in the accompanying figures. For simplicity and clarity of illustration, elements illustrated in the figures are not necessarily drawn to scale. Where considered appropriate, reference labels have been repeated among the figures to indicate corresponding or analogous elements.

FIG. 1 is a diagram of a conceptual overview of a data center in which one or more techniques described herein may be implemented according to various embodiments;

FIG. 2 is a diagram of an example embodiment of a logical configuration of a rack of the data center of FIG. 1;

FIG. 3 is a diagram of an example embodiment of another data center in which one or more techniques described herein may be implemented according to various embodiments;

FIG. 4 is a diagram of another example embodiment of a data center in which one or more techniques described herein may be implemented according to various embodiments;

FIG. 5 is a diagram of a connectivity scheme representative of link-layer connectivity that may be established among various sleds of the data centers of FIGS. 1, 3, and 4;

FIG. 6 is a diagram of a rack architecture that may be representative of an architecture of any particular one of the racks depicted in FIGS. 1-4 according to some embodiments;

FIG. 7 is a diagram of an example embodiment of a sled that may be used with the rack architecture of FIG. 6;

FIG. 8 is a diagram of an example embodiment of a rack architecture to provide support for sleds featuring expansion capabilities;

FIG. 9 is a diagram of an example embodiment of a rack implemented according to the rack architecture of FIG. 8;

FIG. 10 is a diagram of an example embodiment of a sled designed for use in conjunction with the rack of FIG. 9;

FIG. 11 is a diagram of an example embodiment of a data center in which one or more techniques described herein may be implemented according to various embodiments;

FIG. 12 is a simplified block diagram of at least one embodiment of a system for performing data deduplication in a disaggregated architecture;

FIG. 13 is a simplified block diagram of at least one embodiment of a network switch of the system of FIG. 12;

FIG. 14 is a simplified block diagram of at least one embodiment of an environment that may be established by the network switch of FIGS. 12 and 13; and

FIGS. 15-18 are a simplified flow diagram of at least one embodiment of a method for performing data deduplication that may be performed by the network switch of FIGS. 12 and 13.

DETAILED DESCRIPTION OF THE DRAWINGS

While the concepts of the present disclosure are susceptible to various modifications and alternative forms, specific embodiments thereof have been shown by way of example in the drawings and will be described herein in detail. It should be understood, however, that there is no intent to limit the concepts of the present disclosure to the particular forms disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives consistent with the present disclosure and the appended claims.

References in the specification to “one embodiment,” “an embodiment,” “an illustrative embodiment,” etc., indicate that the embodiment described may include a particular feature, structure, or characteristic, but every embodiment may or may not necessarily include that particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the art to effect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described. Additionally, it should be appreciated that items included in a list in the form of “at least one A, B, and C” can mean (A); (B); (C); (A and B); (A and C); (B and C); or (A, B, and C). Similarly, items listed in the form of “at least one of A, B, or C” can mean (A); (B); (C); (A and B); (A and C); (B and C); or (A, B, and C).

The disclosed embodiments may be implemented, in some cases, in hardware, firmware, software, or any combination thereof. The disclosed embodiments may also be implemented as instructions carried by or stored on a transitory or non-transitory machine-readable (e.g., computer-readable) storage medium, which may be read and executed by one or more processors. A machine-readable storage medium may be embodied as any storage device, mechanism, or other physical structure for storing or transmitting information in a form readable by a machine (e.g., a volatile or non-volatile memory, a media disc, or other media device).

In the drawings, some structural or method features may be shown in specific arrangements and/or orderings. However, it should be appreciated that such specific arrangements and/or orderings may not be required. Rather, in some embodiments, such features may be arranged in a different manner and/or order than shown in the illustrative figures. Additionally, the inclusion of a structural or method feature in a particular figure is not meant to imply that such feature is required in all embodiments and, in some embodiments, may not be included or may be combined with other features.

FIG. 1 illustrates a conceptual overview of a data center 100 that may generally be representative of a data center or other type of computing network in/for which one or more techniques described herein may be implemented according to various embodiments. As shown in FIG. 1, data center 100 may generally contain a plurality of racks, each of which may house computing equipment comprising a respective set of physical resources. In the particular non-limiting example depicted in FIG. 1, data center 100 contains four racks 102A to 102D, which house computing equipment comprising respective sets of physical resources (PCRs) 105A to 105D. According to this example, a collective set of physical resources 106 of data center 100 includes the various sets of physical resources 105A to 105D that are distributed among racks 102A to 102D. Physical resources 106 may include resources of multiple types, such as—for example—processors, co-processors, accelerators, field programmable gate arrays (FPGAs), memory, and storage. The embodiments are not limited to these examples.

The illustrative data center 100 differs from typical data centers in many ways. For example, in the illustrative embodiment, the circuit boards (“sleds”) on which components such as CPUs, memory, and other components are placed for increased thermal performance. In particular, in the illustrative embodiment, the sleds are shallower than typical boards. In other words, the sleds are shorter from the front to the back, where cooling fans are located. This decreases the length of the path that air must to travel across the components on the board. Further, the components on the sled are spaced further apart than in typical circuit boards, and the components are arranged to reduce or eliminate shadowing (i.e., one component in the air flow path of another component). In the illustrative embodiment, processing components such as the processors are located on a top side of a sled while near memory, such as DIMMs, are located on a bottom side of the sled. As a result of the enhanced airflow provided by this design, the components may operate at higher frequencies and power levels than in typical systems, thereby increasing performance Furthermore, the sleds are configured to blindly mate with power and data communication cables in each rack 102A, 102B, 102C, 102D, enhancing their ability to be quickly removed, upgraded, reinstalled, and/or replaced. Similarly, individual components located on the sleds, such as processors, accelerators, memory, and data storage drives, are configured to be easily upgraded due to their increased spacing from each other. In the illustrative embodiment, the components additionally include hardware attestation features to prove their authenticity.

Furthermore, in the illustrative embodiment, the data center 100 utilizes a single network architecture (“fabric”) that supports multiple other network architectures including Ethernet and Omni-Path. The sleds, in the illustrative embodiment, are coupled to switches via optical fibers, which provide higher bandwidth and lower latency than typical twisted pair cabling (e.g., Category 5, Category 5e, Category 6, etc.). Due to the high bandwidth, low latency interconnections and network architecture, the data center 100 may, in use, pool resources, such as memory, accelerators (e.g., graphics accelerators, FPGAs, ASICs, etc.), and data storage drives that are physically disaggregated, and provide them to compute resources (e.g., processors) on an as needed basis, enabling the compute resources to access the pooled resources as if they were local. The illustrative data center 100 additionally receives utilization information for the various resources, predicts resource utilization for different types of workloads based on past resource utilization, and dynamically reallocates the resources based on this information.

The racks 102A, 102B, 102C, 102D of the data center 100 may include physical design features that facilitate the automation of a variety of types of maintenance tasks. For example, data center 100 may be implemented using racks that are designed to be robotically-accessed, and to accept and house robotically-manipulatable resource sleds. Furthermore, in the illustrative embodiment, the racks 102A, 102B, 102C, 102D include integrated power sources that receive a greater voltage than is typical for power sources. The increased voltage enables the power sources to provide additional power to the components on each sled, enabling the components to operate at higher than typical frequencies.

FIG. 2 illustrates an exemplary logical configuration of a rack 202 of the data center 100. As shown in FIG. 2, rack 202 may generally house a plurality of sleds, each of which may comprise a respective set of physical resources. In the particular non-limiting example depicted in FIG. 2, rack 202 houses sleds 204-1 to 204-4 comprising respective sets of physical resources 205-1 to 205-4, each of which constitutes a portion of the collective set of physical resources 206 comprised in rack 202. With respect to FIG. 1, if rack 202 is representative of—for example—rack 102A, then physical resources 206 may correspond to the physical resources 105A comprised in rack 102A. In the context of this example, physical resources 105A may thus be made up of the respective sets of physical resources, including physical storage resources 205-1, physical accelerator resources 205-2, physical memory resources 205-3, and physical compute resources 205-4 comprised in the sleds 204-1 to 204-4 of rack 202. The embodiments are not limited to this example. Each sled may contain a pool of each of the various types of physical resources (e.g., compute, memory, accelerator, storage). By having robotically accessible and robotically manipulatable sleds comprising disaggregated resources, each type of resource can be upgraded independently of each other and at their own optimized refresh rate.

FIG. 3 illustrates an example of a data center 300 that may generally be representative of one in/for which one or more techniques described herein may be implemented according to various embodiments. In the particular non-limiting example depicted in FIG. 3, data center 300 comprises racks 302-1 to 302-32. In various embodiments, the racks of data center 300 may be arranged in such fashion as to define and/or accommodate various access pathways. For example, as shown in FIG. 3, the racks of data center 300 may be arranged in such fashion as to define and/or accommodate access pathways 311A, 311B, 311C, and 311D. In some embodiments, the presence of such access pathways may generally enable automated maintenance equipment, such as robotic maintenance equipment, to physically access the computing equipment housed in the various racks of data center 300 and perform automated maintenance tasks (e.g., replace a failed sled, upgrade a sled). In various embodiments, the dimensions of access pathways 311A, 311B, 311C, and 311D, the dimensions of racks 302-1 to 302-32, and/or one or more other aspects of the physical layout of data center 300 may be selected to facilitate such automated operations. The embodiments are not limited in this context.

FIG. 4 illustrates an example of a data center 400 that may generally be representative of one in/for which one or more techniques described herein may be implemented according to various embodiments. As shown in FIG. 4, data center 400 may feature an optical fabric 412. Optical fabric 412 may generally comprise a combination of optical signaling media (such as optical cabling) and optical switching infrastructure via which any particular sled in data center 400 can send signals to (and receive signals from) each of the other sleds in data center 400. The signaling connectivity that optical fabric 412 provides to any given sled may include connectivity both to other sleds in a same rack and sleds in other racks. In the particular non-limiting example depicted in FIG. 4, data center 400 includes four racks 402A to 402D. Racks 402A to 402D house respective pairs of sleds 404A-1 and 404A-2, 404B-1 and 404B-2, 404C-1 and 404C-2, and 404D-1 and 404D-2. Thus, in this example, data center 400 comprises a total of eight sleds. Via optical fabric 412, each such sled may possess signaling connectivity with each of the seven other sleds in data center 400. For example, via optical fabric 412, sled 404A-1 in rack 402A may possess signaling connectivity with sled 404A-2 in rack 402A, as well as the six other sleds 404B-1, 404B-2, 404C-1, 404C-2, 404D-1, and 404D-2 that are distributed among the other racks 402B, 402C, and 402D of data center 400. The embodiments are not limited to this example.

FIG. 5 illustrates an overview of a connectivity scheme 500 that may generally be representative of link-layer connectivity that may be established in some embodiments among the various sleds of a data center, such as any of example data centers 100, 300, and 400 of FIGS. 1, 3, and 4. Connectivity scheme 500 may be implemented using an optical fabric that features a dual-mode optical switching infrastructure 514. Dual-mode optical switching infrastructure 514 may generally comprise a switching infrastructure that is capable of receiving communications according to multiple link-layer protocols via a same unified set of optical signaling media, and properly switching such communications. In various embodiments, dual-mode optical switching infrastructure 514 may be implemented using one or more dual-mode optical switches 515. In various embodiments, dual-mode optical switches 515 may generally comprise high-radix switches. In some embodiments, dual-mode optical switches 515 may comprise multi-ply switches, such as four-ply switches. In various embodiments, dual-mode optical switches 515 may feature integrated silicon photonics that enable them to switch communications with significantly reduced latency in comparison to conventional switching devices. In some embodiments, dual-mode optical switches 515 may constitute leaf switches 530 in a leaf-spine architecture additionally including one or more dual-mode optical spine switches 520.

In various embodiments, dual-mode optical switches may be capable of receiving both Ethernet protocol communications carrying Internet Protocol (IP packets) and communications according to a second, high-performance computing (HPC) link-layer protocol (e.g., Intel's Omni-Path Architecture's, InfiniBand™) via optical signaling media of an optical fabric. As reflected in FIG. 5, with respect to any particular pair of sleds 504A and 504B possessing optical signaling connectivity to the optical fabric, connectivity scheme 500 may thus provide support for link-layer connectivity via both Ethernet links and HPC links. Thus, both Ethernet and HPC communications can be supported by a single high-bandwidth, low-latency switch fabric. The embodiments are not limited to this example.

FIG. 6 illustrates a general overview of a rack architecture 600 that may be representative of an architecture of any particular one of the racks depicted in FIGS. 1 to 4 according to some embodiments. As reflected in FIG. 6, rack architecture 600 may generally feature a plurality of sled spaces into which sleds may be inserted, each of which may be robotically-accessible via a rack access region 601. In the particular non-limiting example depicted in FIG. 6, rack architecture 600 features five sled spaces 603-1 to 603-5. Sled spaces 603-1 to 603-5 feature respective multi-purpose connector modules (MPCMs) 616-1 to 616-5.

FIG. 7 illustrates an example of a sled 704 that may be representative of a sled of such a type. As shown in FIG. 7, sled 704 may comprise a set of physical resources 705, as well as an MPCM 716 designed to couple with a counterpart MPCM when sled 704 is inserted into a sled space such as any of sled spaces 603-1 to 603-5 of FIG. 6. Sled 704 may also feature an expansion connector 717. Expansion connector 717 may generally comprise a socket, slot, or other type of connection element that is capable of accepting one or more types of expansion modules, such as an expansion sled 718. By coupling with a counterpart connector on expansion sled 718, expansion connector 717 may provide physical resources 705 with access to supplemental computing resources 705B residing on expansion sled 718. The embodiments are not limited in this context.

FIG. 8 illustrates an example of a rack architecture 800 that may be representative of a rack architecture that may be implemented in order to provide support for sleds featuring expansion capabilities, such as sled 704 of FIG. 7. In the particular non-limiting example depicted in FIG. 8, rack architecture 800 includes seven sled spaces 803-1 to 803-7, which feature respective MPCMs 816-1 to 816-7. Sled spaces 803-1 to 803-7 include respective primary regions 803-1A to 803-7A and respective expansion regions 803-1B to 803-7B. With respect to each such sled space, when the corresponding MPCM is coupled with a counterpart MPCM of an inserted sled, the primary region may generally constitute a region of the sled space that physically accommodates the inserted sled. The expansion region may generally constitute a region of the sled space that can physically accommodate an expansion module, such as expansion sled 718 of FIG. 7, in the event that the inserted sled is configured with such a module.

FIG. 9 illustrates an example of a rack 902 that may be representative of a rack implemented according to rack architecture 800 of FIG. 8 according to some embodiments. In the particular non-limiting example depicted in FIG. 9, rack 902 features seven sled spaces 903-1 to 903-7, which include respective primary regions 903-1A to 903-7A and respective expansion regions 903-1B to 903-7B. In various embodiments, temperature control in rack 902 may be implemented using an air cooling system. For example, as reflected in FIG. 9, rack 902 may feature a plurality of fans 921 that are generally arranged to provide air cooling within the various sled spaces 903-1 to 903-7. In some embodiments, the height of the sled space is greater than the conventional “1 U” server height. In such embodiments, fans 921 may generally comprise relatively slow, large diameter cooling fans as compared to fans used in conventional rack configurations. Running larger diameter cooling fans at lower speeds may increase fan lifetime relative to smaller diameter cooling fans running at higher speeds while still providing the same amount of cooling. The sleds are physically shallower than conventional rack dimensions. Further, components are arranged on each sled to reduce thermal shadowing (i.e., not arranged serially in the direction of air flow). As a result, the wider, shallower sleds allow for an increase in device performance because the devices can be operated at a higher thermal envelope (e.g., 250 W) due to improved cooling (i.e., no thermal shadowing, more space between devices, more room for larger heat sinks, etc.).

MPCMs 916-1 to 916-7 may be configured to provide inserted sleds with access to power sourced by respective power modules 920-1 to 920-7, each of which may draw power from an external power source 919. In various embodiments, external power source 919 may deliver alternating current (AC) power to rack 902, and power modules 920-1 to 920-7 may be configured to convert such AC power to direct current (DC) power to be sourced to inserted sleds. In some embodiments, for example, power modules 920-1 to 920-7 may be configured to convert 277-volt AC power into 12-volt DC power for provision to inserted sleds via respective MPCMs 916-1 to 916-7. The embodiments are not limited to this example.

MPCMs 916-1 to 916-7 may also be arranged to provide inserted sleds with optical signaling connectivity to a dual-mode optical switching infrastructure 914, which may be the same as—or similar to—dual-mode optical switching infrastructure 514 of FIG. 5. In various embodiments, optical connectors contained in MPCMs 916-1 to 916-7 may be designed to couple with counterpart optical connectors contained in MPCMs of inserted sleds to provide such sleds with optical signaling connectivity to dual-mode optical switching infrastructure 914 via respective lengths of optical cabling 922-1 to 922-7. In some embodiments, each such length of optical cabling may extend from its corresponding MPCM to an optical interconnect loom 923 that is external to the sled spaces of rack 902. In various embodiments, optical interconnect loom 923 may be arranged to pass through a support post or other type of load-bearing element of rack 902. The embodiments are not limited in this context. Because inserted sleds connect to an optical switching infrastructure via MPCMs, the resources typically spent in manually configuring the rack cabling to accommodate a newly inserted sled can be saved.

FIG. 10 illustrates an example of a sled 1004 that may be representative of a sled designed for use in conjunction with rack 902 of FIG. 9 according to some embodiments. Sled 1004 may feature an MPCM 1016 that comprises an optical connector 1016A and a power connector 1016B, and that is designed to couple with a counterpart MPCM of a sled space in conjunction with insertion of MPCM 1016 into that sled space. Coupling MPCM 1016 with such a counterpart MPCM may cause power connector 1016 to couple with a power connector comprised in the counterpart MPCM. This may generally enable physical resources 1005 of sled 1004 to source power from an external source, via power connector 1016 and power transmission media 1024 that conductively couples power connector 1016 to physical resources 1005.

Sled 1004 may also include dual-mode optical network interface circuitry 1026. Dual-mode optical network interface circuitry 1026 may generally comprise circuitry that is capable of communicating over optical signaling media according to each of multiple link-layer protocols supported by dual-mode optical switching infrastructure 914 of FIG. 9. In some embodiments, dual-mode optical network interface circuitry 1026 may be capable both of Ethernet protocol communications and of communications according to a second, high-performance protocol. In various embodiments, dual-mode optical network interface circuitry 1026 may include one or more optical transceiver modules 1027, each of which may be capable of transmitting and receiving optical signals over each of one or more optical channels. The embodiments are not limited in this context.

Coupling MPCM 1016 with a counterpart MPCM of a sled space in a given rack may cause optical connector 1016A to couple with an optical connector comprised in the counterpart MPCM. This may generally establish optical connectivity between optical cabling of the sled and dual-mode optical network interface circuitry 1026, via each of a set of optical channels 1025. Dual-mode optical network interface circuitry 1026 may communicate with the physical resources 1005 of sled 1004 via electrical signaling media 1028. In addition to the dimensions of the sleds and arrangement of components on the sleds to provide improved cooling and enable operation at a relatively higher thermal envelope (e.g., 250 W), as described above with reference to FIG. 9, in some embodiments, a sled may include one or more additional features to facilitate air cooling, such as a heatpipe and/or heat sinks arranged to dissipate heat generated by physical resources 1005. It is worthy of note that although the example sled 1004 depicted in FIG. 10 does not feature an expansion connector, any given sled that features the design elements of sled 1004 may also feature an expansion connector according to some embodiments. The embodiments are not limited in this context.

FIG. 11 illustrates an example of a data center 1100 that may generally be representative of one in/for which one or more techniques described herein may be implemented according to various embodiments. As reflected in FIG. 11, a physical infrastructure management framework 1150A may be implemented to facilitate management of a physical infrastructure 1100A of data center 1100. In various embodiments, one function of physical infrastructure management framework 1150A may be to manage automated maintenance functions within data center 1100, such as the use of robotic maintenance equipment to service computing equipment within physical infrastructure 1100A. In some embodiments, physical infrastructure 1100A may feature an advanced telemetry system that performs telemetry reporting that is sufficiently robust to support remote automated management of physical infrastructure 1100A. In various embodiments, telemetry information provided by such an advanced telemetry system may support features such as failure prediction/prevention capabilities and capacity planning capabilities. In some embodiments, physical infrastructure management framework 1150A may also be configured to manage authentication of physical infrastructure components using hardware attestation techniques. For example, robots may verify the authenticity of components before installation by analyzing information collected from a radio frequency identification (RFID) tag associated with each component to be installed. The embodiments are not limited in this context.

As shown in FIG. 11, the physical infrastructure 1100A of data center 1100 may comprise an optical fabric 1112, which may include a dual-mode optical switching infrastructure 1114. Optical fabric 1112 and dual-mode optical switching infrastructure 1114 may be the same as—or similar to—optical fabric 412 of FIG. 4 and dual-mode optical switching infrastructure 514 of FIG. 5, respectively, and may provide high-bandwidth, low-latency, multi-protocol connectivity among sleds of data center 1100. As discussed above, with reference to FIG. 1, in various embodiments, the availability of such connectivity may make it feasible to disaggregate and dynamically pool resources such as accelerators, memory, and storage. In some embodiments, for example, one or more pooled accelerator sleds 1130 may be included among the physical infrastructure 1100A of data center 1100, each of which may comprise a pool of accelerator resources—such as co-processors and/or FPGAs, for example—that is globally accessible to other sleds via optical fabric 1112 and dual-mode optical switching infrastructure 1114.

In another example, in various embodiments, one or more pooled storage sleds 1132 may be included among the physical infrastructure 1100A of data center 1100, each of which may comprise a pool of storage resources that is globally accessible to other sleds via optical fabric 1112 and dual-mode optical switching infrastructure 1114. In some embodiments, such pooled storage sleds 1132 may comprise pools of solid-state storage devices such as solid-state drives (SSDs). In various embodiments, one or more high-performance processing sleds 1134 may be included among the physical infrastructure 1100A of data center 1100. In some embodiments, high-performance processing sleds 1134 may comprise pools of high-performance processors, as well as cooling features that enhance air cooling to yield a higher thermal envelope of up to 250 W or more. In various embodiments, any given high-performance processing sled 1134 may feature an expansion connector 1117 that can accept a far memory expansion sled, such that the far memory that is locally available to that high-performance processing sled 1134 is disaggregated from the processors and near memory comprised on that sled. In some embodiments, such a high-performance processing sled 1134 may be configured with far memory using an expansion sled that comprises low-latency SSD storage. The optical infrastructure allows for compute resources on one sled to utilize remote accelerator/FPGA, memory, and/or SSD resources that are disaggregated on a sled located on the same rack or any other rack in the data center. The remote resources can be located one switch jump away or two-switch jumps away in the spine-leaf network architecture described above with reference to FIG. 5. The embodiments are not limited in this context.

In various embodiments, one or more layers of abstraction may be applied to the physical resources of physical infrastructure 1100A in order to define a virtual infrastructure, such as a software-defined infrastructure 1100B. In some embodiments, virtual computing resources 1136 of software-defined infrastructure 1100B may be allocated to support the provision of cloud services 1140. In various embodiments, particular sets of virtual computing resources 1136 may be grouped for provision to cloud services 1140 in the form of software-defined infrastructure (SDI) services 1138. Examples of cloud services 1140 may include—without limitation—software as a service (SaaS) services 1142, platform as a service (PaaS) services 1144, and infrastructure as a service (IaaS) services 1146.

In some embodiments, management of software-defined infrastructure 1100B may be conducted using a virtual infrastructure management framework 1150B. In various embodiments, virtual infrastructure management framework 1150B may be designed to implement workload fingerprinting techniques and/or machine-learning techniques in conjunction with managing allocation of virtual computing resources 1136 and/or SDI services 1138 to cloud services 1140. In some embodiments, virtual infrastructure management framework 1150B may use/consult telemetry data in conjunction with performing such resource allocation. In various embodiments, an application/service management framework 1150C may be implemented in order to provide QoS management capabilities for cloud services 1140. The embodiments are not limited in this context.

Referring now to FIG. 12, a system 1210 for performing data deduplication in a disaggregated architecture may be implemented in accordance with the data centers 100, 300, 400, 1100 described above with reference to FIGS. 1, 3, 4, and 11. In the illustrative embodiment, data deduplication means storing each sub-block (e.g., 64 bytes) of a larger block of data at a different physical address (e.g., in a particular data storage device of a particular data storage sled) and storing references (e.g., pointers) to each sub-block in a deduplication table. As such, when another block of data is to be written, identical sub-blocks from the subsequent block are not stored in other physical memory locations. Rather, references to the already-stored sub-blocks (e.g., 64 bytes) are stored in the deduplication table in association with the matching sub-blocks of the subsequent data block.

In the illustrative embodiment, the system 1210 includes an orchestrator server 1216 in communication with a network switch 1220. The network switch 1220 is communicatively coupled to multiple sleds including compute sleds 1230, 1232, data storage sleds 1240, 1242, and accelerator sleds 1260, 1262. One or more of the sleds 1230, 1232, 1240, 1242, 1260, 1262 may be grouped into a managed node, such as by the orchestrator server 1216, to collectively perform a workload, such as an application. A managed node may be embodied as an assembly of resources (e.g., physical resources 206), such as compute resources (e.g., physical compute resources 205-4), memory resources (e.g., physical memory resources 205-3), storage resources (e.g., physical storage resources 205-1), or other resources (e.g., physical accelerator resources 205-2), from the same or different sleds (e.g., the sleds 204-1, 204-2, 204-3, 204-4, etc.) or racks (e.g., one or more of racks 302-1 through 302-32). Further, a managed node may be established, defined, or “spun up” by the orchestrator server 1216 at the time a workload is to be assigned to the managed node or at any other time, and may exist regardless of whether any workloads are presently assigned to the managed node. The system 1210 may be located in a data center and provide storage and compute services (e.g., cloud services) to a client device 1214 that is in communication with the system 1210 through a network 1212. The orchestrator server 1216 may support a cloud operating environment, such as OpenStack, and managed nodes established by the orchestrator server 1216 may execute one or more applications or processes (i.e., workloads), such as in virtual machines or containers, on behalf of a user of the client device 1214. In the illustrative embodiment, the compute sled 1230 executes a workload 1234 (e.g., an application), and the compute sled 1232 executes another workload 1236 (e.g., another application). Further, the data storage sled 1240 includes multiple data storage devices 1244, 1246 (e.g., physical storage resources 205-1). Likewise, the data storage sled 1242 includes multiple data storage devices 1248, 1250 (e.g., physical storage resources 205-1). Additionally, the accelerator sled 1260 includes one or more accelerator devices 1264 (e.g., physical accelerator resources 205-2) and the accelerator sled 1266 also includes one or more accelerator devices 1266 (e.g., physical accelerator resources 205-2).

In operation, the system 1210 may utilize one or more deduplication logic units 1270, which may be present in the network switch 1220, one or more of the compute sleds 1230, 1232, and/or the accelerator sleds 1260, 1262, all of which may be referred herein as network devices, to perform deduplication of data across the fabric (e.g., across the sleds in the system 1210). In the illustrative embodiment, the deduplication logic unit 1270 is included in the network switch 1220, and in operation, the network switch 1220 collects and analyzes telemetry data indicative of performance conditions of the sleds as workloads are being performed and information about network traffic passing through the network switch 1220, including network congestion information, frequencies of data access requests and responses between particular compute sleds 1230, 1232 and corresponding storage sleds 1240, 1242, to select where data is to be stored to enable efficient deduplication of the data as the workloads are performed. Additionally, the network switch 1220 may perform a migration of data from one storage sled 1242 to another storage sled 1240 to store, on the same data storage sled, data sub-blocks that are frequently accessed by a particular compute sled 1240 and/or perform load balancing among the data storage sleds 1240, 1242, as described in more detail herein. In some embodiments, the network switch 1220 may utilize a private key corresponding to a public key selected by a compute sled 1230, 1232, to perform cryptographic operations on data sent through the network switch 1220 to and/or from a data storage sled 1240, 1242.

Referring now to FIG. 13, the network switch 1220 may be embodied as any type of compute device capable of performing the functions described herein, including routing data communications among the sleds 1230, 1232, 1240, 1242, 1260, 1262, and/or the orchestrator server 1216, collecting telemetry data indicative of performance conditions of the sleds as the workloads are executed, and providing efficient deduplication of data across the system 1210 (e.g., across multiple disaggregated data storage sleds 1240, 1242) using the telemetry data to determine where to store data sub-blocks (e.g., to locate sub-blocks that are frequently accessed as a set on the same data storage sled, to balance I/O requests across the data storage sleds, etc.).

As shown in FIG. 13, the illustrative network switch 1220 includes a compute engine 1302, an input/output (I/O) subsystem 1308, communication circuitry 1310, and one or more data storage devices 1314. Of course, in other embodiments, the network switch 1220 may include other or additional components, such as those commonly found in a computer (e.g., display, peripheral devices, etc.). Additionally, in some embodiments, one or more of the illustrative components may be incorporated in, or otherwise form a portion of, another component.

The compute engine 1302 may be embodied as any type of device or collection of devices capable of performing various compute functions described below. In some embodiments, the compute engine 1302 may be embodied as a single device such as an integrated circuit, an embedded system, a field-programmable gate array (FPGA), a system-on-a-chip (SOC), or other integrated system or device. Additionally, in some embodiments, the compute engine 1302 includes or is embodied as a processor 1304 and a memory 1306. The processor 1304 may be embodied as any type of processor capable of performing the functions described herein. For example, the processor 1304 may be embodied as a single or multi-core processor(s), a microcontroller, or other processor or processing/controlling circuit. In some embodiments, the processor 1304 may be embodied as, include, or be coupled to an FPGA, an application specific integrated circuit (ASIC), reconfigurable hardware or hardware circuitry, or other specialized hardware to facilitate performance of the functions described herein. The processor 1304 may include the deduplication logic unit 1270 described with reference to FIG. 12. The deduplication logic unit 1270 may be embodied as a specialized device, such as a co-processor, an FPGA, or an ASIC, for performing the deduplication operations described above (e.g., collecting and analyzing telemetry data indicative of performance conditions of the sleds as workloads are being performed and utilizing the telemetry data to select where data is to be stored to enable efficient deduplication of the data as the workloads are performed).

The main memory 1306 may be embodied as any type of volatile (e.g., dynamic random access memory (DRAM), etc.) or non-volatile memory or data storage capable of performing the functions described herein. Volatile memory may be a storage medium that requires power to maintain the state of data stored by the medium. Non-limiting examples of volatile memory may include various types of random access memory (RAM), such as dynamic random access memory (DRAM) or static random access memory (SRAM). One particular type of DRAM that may be used in a memory module is synchronous dynamic random access memory (SDRAM). In particular embodiments, DRAM of a memory component may comply with a standard promulgated by JEDEC, such as JESD79F for DDR SDRAM, JESD79-2F for DDR2 SDRAM, JESD79-3F for DDR3 SDRAM, JESD79-4A for DDR4 SDRAM, JESD209 for Low Power DDR (LPDDR), JESD209-2 for LPDDR2, JESD209-3 for LPDDR3, and JESD209-4 for LPDDR4 (these standards are available at www.jedec.org). Such standards (and similar standards) may be referred to as DDR-based standards and communication interfaces of the storage devices that implement such standards may be referred to as DDR-based interfaces.

In one embodiment, the memory device is a block addressable memory device, such as those based on NAND or NOR technologies. A memory device may also include future generation nonvolatile devices, such as a three dimensional crosspoint memory device, or other byte addressable write-in-place nonvolatile memory devices. In one embodiment, the memory device may be or may include memory devices that use chalcogenide glass, multi-threshold level NAND flash memory, NOR flash memory, single or multi-level Phase Change Memory (PCM), a resistive memory, nanowire memory, ferroelectric transistor random access memory (FeTRAM), anti-ferroelectric memory, magnetoresistive random access memory (MRAM) memory that incorporates memristor technology, resistive memory including the metal oxide base, the oxygen vacancy base and the conductive bridge Random Access Memory (CB-RAM), or spin transfer torque (STT)-MRAM, a spintronic magnetic junction memory based device, a magnetic tunneling junction (MTJ) based device, a DW (Domain Wall) and SOT (Spin Orbit Transfer) based device, a thyristor based memory device, or a combination of any of the above, or other memory. The memory device may refer to the die itself and/or to a packaged memory product.

In some embodiments, 3D crosspoint memory (e.g., Intel 3D XPoint™ memory) may comprise a transistor-less stackable cross point architecture in which memory cells sit at the intersection of word lines and bit lines and are individually addressable and in which bit storage is based on a change in bulk resistance. In some embodiments, all or a portion of the main memory 1306 may be integrated into the processor 1304. In operation, the main memory 1306 may store various software and data used during operation such as telemetry data, deduplication data, migration policy data, key data, applications, programs, libraries, and drivers.

The compute engine 1302 is communicatively coupled to other components of the network switch 1220 via the I/O subsystem 1308, which may be embodied as circuitry and/or components to facilitate input/output operations with the compute engine 1302 (e.g., with the processor 1304 and/or the main memory 1306) and other components of the network switch 1220. For example, the I/O subsystem 1308 may be embodied as, or otherwise include, memory controller hubs, input/output control hubs, integrated sensor hubs, firmware devices, communication links (e.g., point-to-point links, bus links, wires, cables, light guides, printed circuit board traces, etc.), and/or other components and subsystems to facilitate the input/output operations. In some embodiments, the I/O subsystem 1308 may form a portion of a system-on-a-chip (SoC) and be incorporated, along with one or more of the processor 1304, the main memory 1306, and other components of the network switch 1220, into the compute engine 1302.

The communication circuitry 1310 may be embodied as any communication circuit, device, or collection thereof, capable of enabling communications over the network 1212 between the network switch 1220 and another compute device (e.g., the orchestrator server 1216, and/or one or more sleds 1230, 1232, 1240, 1242, 1260, 1262). The communication circuitry 1310 may be configured to use any one or more communication technology (e.g., wired or wireless communications) and associated protocols (e.g., Ethernet, Bluetooth®, Wi-Fi®, WiMAX, etc.) to effect such communication.

The illustrative communication circuitry 1310 includes one or more port logics 1312. Each port logic 1312 may be embodied as one or more add-in-boards, daughter cards, network interface cards, controller chips, chipsets, or other devices that may be used by the network switch 1220 to connect with another compute device (e.g., the orchestrator server 1216 and/or the sleds 1230, 1232, 1240, 1242, 1260, 1262). In some embodiments, the one or more port logics 1312 may be embodied as part of a system-on-a-chip (SoC) that includes one or more processors, or included on a multichip package that also contains one or more processors. In some embodiments, the one or more port logics 1312 may include a local processor (not shown) and/or a local memory (not shown) that are both local to the port logic(s) 1312. In such embodiments, the local processor of the port logic(s) 1312 may be capable of performing one or more of the functions of the compute engine 1302 described herein. Additionally or alternatively, in such embodiments, the local memory of the port logic(s) 1312 may be integrated into one or more components of the network switch 1220 at the board level, socket level, chip level, and/or other levels. In some embodiments, the deduplication logic unit 1270 may be included in the port logic(s) 1312.

The one or more illustrative data storage devices 1314, may be embodied as any type of devices configured for short-term or long-term storage of data such as, for example, memory devices and circuits, memory cards, hard disk drives, solid-state drives, or other data storage devices. Each data storage device 1314 may include a system partition that stores data and firmware code for the data storage device 1314. Each data storage device 1314 may also include an operating system partition that stores data files and executables for an operating system.

Additionally or alternatively, the network switch 1220 may include one or more peripheral devices 1316. Such peripheral devices 1316 may include any type of peripheral device commonly found in a compute device such as a display, speakers, a mouse, a keyboard, and/or other input/output devices, interface devices, and/or other peripheral devices.

The client device 1214, the orchestrator server 1216, and the sleds 1230, 1232, 1240, 1242, 1260, 1262 may have components similar to those described in FIG. 13. The description of those components of the network switch 1220 is equally applicable to the description of components of those devices and is not repeated herein for clarity of the description. Further, it should be appreciated that any of the client device 1214, the orchestrator server 1216, and the sleds 1230, 1232, 1240, 1242, 1260, 1262 may include other components, sub-components, and devices commonly found in a computing device, which are not discussed above in reference to the network switch 1220 and not discussed herein for clarity of the description.

As described above, the network switch 1220, the orchestrator server 1216, and the sleds 1230, 1232, 1240, 1242, 1260, 1262 are illustratively in communication via the network 1212, which may be embodied as any type of wired or wireless communication network, including global networks (e.g., the Internet), local area networks (LANs) or wide area networks (WANs), cellular networks (e.g., Global System for Mobile Communications (GSM), 3G, Long Term Evolution (LTE), Worldwide Interoperability for Microwave Access (WiMAX), etc.), digital subscriber line (DSL) networks, cable networks (e.g., coaxial networks, fiber networks, etc.), or any combination thereof.

Referring now to FIG. 14, the network switch 1220 may establish an environment 1400 during operation. The illustrative environment 1400 includes a network communicator 1420, a telemetry data collector 1430, and a deduplication manager 1440. Each of the components of the environment 1400 may be embodied as hardware, firmware, software, or a combination thereof. As such, in some embodiments, one or more of the components of the environment 1400 may be embodied as circuitry or a collection of electrical devices (e.g., network communicator circuitry 1420, telemetry data collector circuitry 1430, deduplication manager circuitry 1440, etc.). It should be appreciated that, in such embodiments, one or more of the network communicator circuitry 1420, telemetry data collector circuitry 1430, or deduplication manager circuitry 1440 may form a portion of one or more of the compute engine 1302, the deduplication logic unit 1270, the communication circuitry 1310, the I/O subsystem 1308, and/or other components of the network switch 1220. In the illustrative embodiment, the environment 1400 includes telemetry data 1402, which may be embodied as any data collected by the network switch 1220 during the execution of one or more workloads by the sleds 1230, 1232, 1240, 1242, 1260, 1262 and indicative of performance conditions (e.g., load information such as I/O operations per second, percentage of available compute capacity presently used, percentage of available network communication capacity present used, etc.) of the sleds as workloads are being executed. The telemetry data 1402 may also include information about network traffic passing through the network switch 1220, including network congestion information and frequencies of data access requests and responses between particular compute sleds 1230, 1232 and corresponding storage sleds 1240, 1242. Additionally, the illustrative environment 1400 includes deduplication data 1404 which may be embodied as any data indicative of identifiers of sub-blocks of data (e.g., 64 bytes of data) and corresponding pointers to physical addresses (e.g., identifiers of data storage sleds in combination with identifiers of addresses within data storage devices located on the data storage sleds, where each sub-block is stored). In some embodiments, the deduplication data 1404 may include a hash of each data sub-block in association with the corresponding pointer. Further, in the illustrative embodiment, the pointers associated with each sub-block correspond with logical addresses in a logical address space utilized (e.g., included in memory access requests) by the compute sleds 1230, 1232 and/or other sleds during the execution of workloads.

Additionally, in the illustrative embodiment, the environment 1400 includes migration policy data 1406, which may be embodied as any data indicative of rules usable to determine when data is to be migrated from one or more data storage sleds (e.g., the data storage sled 1242) to another data storage sled (e.g., the data storage sled 1240), such as to move data sub-blocks that are frequently accessed contemporaneously or by a particular phase of a workload to be on the same data storage sled (e.g., the data storage sled 1240), to balance I/O loads across the data storage sleds 1240, 1242, and/or to achieve other goals which may be set by an administrator or provided from another source. Additionally, the environment 1400 may include key data 1408, which may be embodied as any data indicative of a set of keys usable to perform cryptographic operations (e.g., encryption and/or decryption) of data sent through the network switch 1220 (e.g., from a compute sled 1230 to a data storage sled 1240 or vice versa). In the illustrative embodiment, the key data 1408 includes multiple public keys and corresponding private keys (e.g., one public key and private key pair for each workload executed in the system 1210).

In the illustrative environment 1400, the network communicator 1420, which may be embodied as hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof as discussed above, is configured to facilitate inbound and outbound network communications (e.g., network traffic, network packets, network flows, etc.) to and from the network switch 1220, respectively. To do so, the network communicator 1420 is configured to receive and process data packets from one system or computing device (e.g., a compute sled 1230, 1232) and to prepare and send data packets to another computing device or system (e.g., a data storage sled 1240, 1242). Accordingly, in some embodiments, at least a portion of the functionality of the network communicator 1420 may be performed by the communication circuitry 1310, and, in the illustrative embodiment, by the port logic(s) 1312.

The telemetry data collector 1430, which may be embodied as hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof, is configured to collect telemetry data (e.g., the telemetry data 1402) reported by the sleds 1230, 1232, 1240, 1242, 1260, 1262 as the workloads are executed. In some embodiments, the telemetry data 1402 may be destined for the orchestrator server 1216 and, as packets containing the telemetry data 1402 pass through the network switch 1220 (e.g., through the network communicator 1420), the telemetry data collector 1430 identifies those packets, and stores the telemetry data 1402 locally in the network switch 1220 in association with each corresponding sled.

The deduplication manager 1440, which may be embodied as hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof, is configured to provide efficient deduplication of data across the system 1210. To do so, in the illustrative embodiment, the deduplication manager 1440 includes a data request manager 1442, a migration manager 1444, and a cryptographic operation manager 1446. The data request manager 1442, in the illustrative embodiment, is configured to respond to data write requests (e.g., from a compute sled 1230) by partitioning a data block that is to be written into multiple sub-blocks (e.g., 64 bytes each), determining whether each sub-block has already been stored in a data storage device in one of the data storage sleds 1240, 1242, storing a pointer (e.g., physical address) to each already-stored sub-block in association with a corresponding logical address (e.g., in the deduplication data 1404), storing new sub-blocks at physical addresses that are not presently used, and storing pointers to those new sub-blocks (e.g., in the deduplication data 1404). In some embodiments, the data request manager 1442 may generate a hash of each sub-block, store the hash in association with each corresponding pointer, and when determining whether a particular sub-block is identical to an already-stored sub-block, first compare the hashes to determine whether they are identical, and if so, compare the actual sub-blocks to each other. In response to a read request, the data request manager 1442, in the illustrative embodiment, accesses the deduplication data 1404 associated with a logical address (e.g., a starting address) of a data block to be read, identifies the pointers to the sub-blocks, requests the data sub-blocks from the data storage devices on the corresponding data storage sleds 1240, 1242 associated with the pointers, and sends the data sub-blocks to the requesting sled (e.g., the compute sled 1230, 1232) in response to the read request.

The migration manager 1444, in the illustrative embodiment, is configured to migrate sub-blocks of data from a data storage sled 1242 to another data storage sled 1240 (e.g., by issuing read requests and corresponding write requests to the respective data storage sleds 1242, 1240) in accordance with the migration policy data 1406 described above. The cryptographic operation manager 1446, in the illustrative embodiment, is configured to perform cryptographic operations on data sent through the network switch 1220 on behalf of a workload executed by a compute sled 1230, 1232. In the illustrative embodiment, the cryptographic operation manager 1446 performs the cryptographic operations using a different key for each workload (e.g., a private key corresponding to a public key selected by a compute sled 1230, 1232 executing a particular workload). For example, and as described in more detail herein, the cryptographic operation manager 1446 may encrypt sub-blocks of data submitted by a compute sled 1230 in a write request using a private key (e.g., in the key data 1408) that corresponds with a public key (e.g., also in the key data 1408) selected by the compute sled 1230.

It should be appreciated that each of the data request manager 1442, the migration manager 1444, and the cryptographic operation manager 1446 may be separately embodied as hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof. For example, the data request manager 1442 may be embodied as a hardware component, while the migration manager 1444 and the cryptographic operation manager 1446 are embodied as virtualized hardware components or as some other combination of hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof. Further it should be appreciated that in some embodiments, a sled 1230, 1232, 1260, 1262 containing the deduplication logic unit 1270 may establish an environment similar to the environment 1400 described above.

Referring now to FIG. 15, in use, a network device (e.g., the network switch 1220 and/or a sled 1230, 1232, 1260, 1262 containing the deduplication logic unit 1270) may execute a method 1500 for performing deduplication of data. For simplicity, the method 1500 is described below as being performed by the network switch 1220. However, it should be understood that in other embodiments, the method 1500 may be performed by one or more other network devices (e.g., a sled 1230, 1232, 1260, 1262 containing the deduplication logic unit 1270). The method 1500 begins with block 1502 in which the network switch 1220 determines whether to enable data deduplication in a disaggregated architecture (e.g., the system 1210). In the illustrative embodiment, the network switch 1220 determines to enable data deduplication in response to determining that a configuration file (e.g., stored in a data storage device 1314) includes a setting indicating that data deduplication should be enabled, in response to a request from an administrator compute device (not shown), and/or based on other factors. Regardless, in response to a determination to enable data deduplication, the method 1500 advances to block 1504 in which the network switch 1220, in the illustrative embodiment, collects telemetry data (e.g., the telemetry data 1402) indicative of performance conditions of sleds (e.g., the sleds 1230, 1232, 1240, 1242, 1260, 1262) connected to the network switch 1220. In doing so, the network switch 1220 may collect telemetry data indicative of a load on each data storage sled 1240, 1242 (e.g., a number of I/O operations per second, network congestion experienced by each data storage sled 1240, 1242, etc.), as indicated in block 1506. As indicated in block 1508, in collecting the telemetry data 1402, the network switch 1220 may collect telemetry data 1402 indicative of a frequency of data access operations (e.g., reads, writes, etc.), the data accessed (e.g., logical addresses, physical addresses, hashes of the data, etc.), and identifiers (e.g., internet protocol (IP) addresses, media access control addresses (MAC), etc.) and locations (e.g., port numbers which may be correlated to corresponding racks and/or sleds within a rack).

Subsequently, in block 1510, the network switch 1220 may receive, from a compute sled (e.g., the compute sled 1230), a selection of a public key associated with a private key stored by the network switch 1220 (e.g., in the key data 1408). In block 1512, the network switch 1220 receives, from a compute sled (e.g., the compute sled 1230) a request to perform a data access operation. In doing so, the network switch 1220 may receive a write request, as indicated in block 1514. The write request, in the illustrative embodiment, includes a payload of data to be written to a data storage sled (e.g., the data storage sled 1240). Further, and as indicated in block 1516, in receiving the write request, the network switch 1220 may receive a write request with encrypted data, such as data encrypted with the public key selected in block 1510. Alternatively, the request may be a read request that includes a logical address from which to read data (e.g., a block of data), as indicated in block 1518. In block 1520, the network switch 1220 determines the subsequent actions to take as a function of whether a write request was received in block 1512. If the request is a write request, the method 1500 advances to block 1522 of FIG. 16, in which the network switch 1220 determines, with the deduplication data 1404, whether any sub-blocks within the block of data to be written have already been written to a data storage device of one of the data storage sleds 1240, 1242. Otherwise, the method 1500 advances to block 1546 of FIG. 17, in which the network switch 1220 determines whether a read request was instead received in block 1520.

Referring now to FIG. 16, in block 1522, the network switch 1220 determines, with the deduplication data 1404, whether each of multiple sub-blocks (e.g., 64 bytes of data) of the data to be written have already been written to a data storage device of a data storage sled 1240, 1242. In doing so, in block 1524, the network switch 1220 may decrypt the data with a key associated with the key selected by the compute sled 1230 (e.g., a private key associated with the public key selected in block 1510 and stored in the key data 1408). In block 1526, the network switch 1220 may perform the determination with a local deduplication logic unit (e.g., the deduplication logic unit 1270 included in the network switch 1220). In other embodiments, the network switch 1220 may perform the determination with a remote deduplication logic unit 1270, as indicated in block 1528. For example, and as indicated in block 1530, the network switch 1220 may perform the determination with a deduplication logic unit 1270 of an accelerator sled (e.g., the accelerator sled 1260), such as by sending the data to be written to the accelerator sled 1260 for analysis and receiving results of the analysis from the accelerator sled 1260 in response. Regardless, in the illustrative embodiment, the determination is performed by generating (e.g., with the deduplication logic unit 1270) a hash of each sub-block, as indicated in block 1532. Additionally, and as indicated in block 1534, a determination is made (e.g., with the deduplication logic unit 1270) of whether data has been stored in a data storage device of a data storage sled 1240, 1242 in association with the generated hash (e.g., by comparing the generated hash to hashes in the deduplication data 1404). Further, a determination is made as to whether data stored in association with the hash matches (e.g., is identical to the present sub-block), as indicated in block 1536. In the illustrative embodiment, if the generated hash of a sub-block matches a hash that is present in the deduplication data 1404, the network switch 1220 then identifies the physical address associated with the matching hash, reads the sub-block at the physical address (e.g., by requesting the sub-block from the corresponding data storage sled 1240, 1242) and performs a byte-by-byte comparison of the sub-blocks to determine whether the sub-blocks are identical.

Subsequently, in block 1538, the network switch 1220 writes the data and updates the deduplication data 1404. In doing so, and as indicated in block 1540, the network switch 1220 writes, in association with a logical address, a pointer to a physical address of an existing stored sub-block for any duplicate sub-block (e.g., any sub-block that matches an already-stored sub-block). As indicated in block 1542, the network switch 1220 writes non-duplicative sub-blocks (e.g., the sub-blocks that do not match already-stored sub-blocks) to unused physical addresses (e.g., locations in data storage devices of the data storage sleds 1240, 1242) and stores the physical addresses in association with the hashes and corresponding logical addresses in the deduplication data 1404. In doing so, and as indicated in block 1544, the network switch 1220 may select a physical address for each sub-block as a function of data storage sled loads and location data in the collected telemetry data 1402. For example, the network switch 1220 may determine to store data sub-blocks across multiple data storage sleds 1240, 1242 to balance I/O loads and/or network congestion. Additionally or alternatively, the network switch 1220 may determine to store data sub-blocks on a data storage sled 1240 that is in the same rack as the compute sled 1230 that sent the write request. Subsequently, the method 1500 advances to block 1566 of FIG. 18 in which the network switch 1220 determines whether to migrate data from one data storage sled 1240 to another data storage sled 1242.

Referring back to block 1520 of FIG. 15, if the network switch 1220 instead determines that a write request was not received, the method 1500 advances to block 1546 of FIG. 17, in which the network switch 1220 determines whether a read request was received. Referring now to FIG. 17, if the network switch 1220 determines that a read request was received, the method 1500 advances to block 1548, in which the network switch 1220 references the deduplication data 1404 to identify physical addresses of sub-blocks within the block to be read. In doing so, the network switch 1220 may identify the physical addresses with a local deduplication logic unit (e.g., the deduplication logic unit 1270 in the network switch 1220), as indicated in block 1550. Alternatively, and as indicated in block 1552, the network switch 1220 may identify the physical addresses with a remote deduplication logic unit 1270. For example, and as indicated in block 1554, the network switch 1220 may identify the physical addresses with a remote deduplication logic unit 1270 in an accelerator sled (e.g., the accelerator sled 1260), such as by sending a request to the accelerator sled 1260 to perform the identification operation and receiving, from the accelerator sled 1260, the physical addresses.

Subsequently, in block 1556, the network switch 1220 accesses the requested data from the corresponding physical addresses. In doing so, in the illustrative embodiment, the network switch 1220 accesses the requested data from data storage devices on one or more data storage sleds 1240, 1242 referenced in the deduplication data 1404 (e.g., using the physical addresses in the deduplication data 1404), as indicated in block 1558. Afterwards, and as indicated in block 1560, the network switch 1220 provides the accessed data to the compute sled 1230 that sent the request in block 1512. In doing so, the switch 1220 may perform a cryptographic operation on the accessed data, as indicated in block 1562. For example, and as indicated in block 1564, the network switch 1220 may encrypt the data with a key associated with the key selected by the compute sled 1230 (e.g., the private key corresponding to a public key selected by the compute sled 1230 in block 1510). Subsequently, or if the network switch 1220 determined that a read request was not received in block 1546, the method 1500 advances to block 1566 of FIG. 18, in which the network switch 1220 determines, as a function of the telemetry data 1402, whether to migrate data from one data storage sled 1242 to another data storage sled 1240.

Referring now to FIG. 18, in determining whether to migrate data, the network switch 1220 may determine whether to migrate data sub-blocks that are frequently (e.g., above a predefined frequency) accessed in a set (e.g., concurrently, or within a predefined time period of each other) to the same data storage sled (e.g., the data storage sled 1240), as indicated in block 1568. As indicated in block 1570, the network switch 1220 may determine whether to migrate data to balance loads across multiple data storage sleds (e.g., the data storage sleds 1240, 1242). In block 1572, the network switch 1220 may determine whether to migrate data to balance network traffic congestion across the data storage sleds 1240, 1242. Additionally or alternatively, as indicated in block 1574, the network switch 1220 may determine whether to migrate data to balance I/O loads and/or wear on the data storage devices of the data storage sleds 1240, 1242. In some embodiments, the network switch 1220 may determine to balance the loads if the loads differ between the data storage sleds 1240, 1242 by at least a threshold amount (e.g., 25%). In other embodiments, the network switch 1220 determine to balance the loads based on other criteria.

In block 1576, the network switch 1220 determines the subsequent actions to perform as a function of whether the network switch 1220 determined to migrate data. If not, the method 1500 loops back to block 1502 of FIG. 15 in which the network switch 1220 determines whether to continue to perform data deduplication in the system 1210. Otherwise, the method 1500 advances to block 1578 in which the network switch 1220 migrates the data (e.g., by issuing read and write requests to the corresponding data storage sleds 1240, 1242) and updates the deduplication data 1404 with modified physical addresses associated with each sub-block of migrated data (e.g., replacing a physical address associated with a data storage device in the data storage sled 1242 where a sub-block was migrated from with a physical address associated with a data storage device in the data storage sled 1240 where the sub-block was migrated to). Subsequently, the method 1500 loops back to block 1502 of FIG. 15, in which the network switch 1220 determines whether to continue to perform data deduplication in the system 1210.

EXAMPLES

Illustrative examples of the technologies disclosed herein are provided below. An embodiment of the technologies may include any one or more, and any combination of, the examples described below.

Example 1 includes a network device comprising communication circuitry to receive, from a compute sled, a request to write a data block to one or more data storage sleds; and a compute engine to determine, for each of one or more data sub-blocks within the data block and from deduplication data indicative of physical addresses of data sub-blocks, whether each data sub-block is already stored in a data storage device of a data storage sled; write, in the deduplication data and in response to a determination that a data sub-block is already stored in a data storage device, a pointer to a physical address of the already-stored data sub-block in association with a logical address of the data sub-block; and write, to a physical address associated with a data storage device in a data storage sled and in response to a determination that a data sub-block is not already stored in a data storage device, the data sub-block and update the deduplication data with a pointer to the physical address in association with the logical address.

Example 2 includes the subject matter of Example 1, and wherein to determine whether each data sub-block is already stored in a data storage device of a data storage sled comprises to generate a hash of each data sub-block; determine, from hashes associated with pointers to physical addresses present in the deduplication data, whether data has been stored in a data storage sled in association with the generated hash; and compare, in response to a determination that data has been stored in a data storage sled in association with the generated hash, the stored data to the data sub-block to determine whether the stored data matches the data sub-block.

Example 3 includes the subject matter of any of Examples 1 and 2, and wherein the compute engine is further to collect telemetry data indicative of the performance and conditions of sleds connected to the network device.

Example 4 includes the subject matter of any of Examples 1-3, and wherein to write a pointer to a physical address comprises to select a physical address as a function of the collected telemetry data.

Example 5 includes the subject matter of any of Examples 1-4, and wherein to select a physical address as a function of the collected telemetry data comprises to select a physical address as a function of data storage sled load data and data storage location data in the collected telemetry data.

Example 6 includes the subject matter of any of Examples 1-5, and wherein the compute engine is further to determine whether a set of data sub-blocks that are distributed across multiple data storage sleds are accessed by a particular compute sled with least at a predefined frequency; and migrate, in response to a determination that the compute sled accesses the data sub-blocks with at least the predefined frequency, the data sub-blocks to a single data storage sled.

Example 7 includes the subject matter of any of Examples 1-6, and wherein the compute engine is further to determine, from the collected telemetry data, whether to migrate data sub-blocks from a first storage sled to a second storage sled to balance loads across the data storage sleds.

Example 8 includes the subject matter of any of Examples 1-7, and wherein the compute engine is further to migrate the data sub-blocks with a function to balance wear across the data storage devices in the data storage sleds.

Example 9 includes the subject matter of any of Examples 1-8, and wherein the compute engine is further to migrate the data sub-blocks with a function to balance I/O loads across the data storage devices in the data storage sleds.

Example 10 includes the subject matter of any of Examples 1-9, and wherein the compute engine is further to migrate the data sub-blocks from a first data storage sled to a second data storage sled with less network congestion than the first data storage sled.

Example 11 includes the subject matter of any of Examples 1-10, and wherein the compute engine is further to receive, from the compute sled, a selection of a key; and perform a cryptographic operation on one or more of the data sub-blocks as a function of the selection of the key.

Example 12 includes the subject matter of any of Examples 1-11, and wherein to determine whether each data sub-block is already stored in a data storage device of a data storage sled comprises to perform the determination with a deduplication logic unit that is local to the network device.

Example 13 includes the subject matter of any of Examples 1-12, and wherein to determine whether each data sub-block is already stored in a data storage device of a data storage sled comprises to perform the determination with a deduplication logic unit that is remote from the network device.

Example 14 includes the subject matter of any of Examples 1-13, and wherein to perform the determination with a deduplication logic unit that is remote from the network device comprises to perform the determination with a deduplication logic unit that is located on an accelerator sled that is communicatively coupled to the network device.

Example 15 includes the subject matter of any of Examples 1-14, and wherein the compute engine is further to receive, from the compute sled, a read request for a data block to be read; reference the deduplication data to identify physical addresses of data sub-blocks within the data block to be read; access the data sub-blocks from the identified physical addresses; and provide the accessed data sub-blocks to the compute sled.

Example 16 includes the subject matter of any of Examples 1-15, and wherein the compute engine is further to collect telemetry data indicative of the performance and conditions of sleds connected to the network device; determine whether a set of data sub-blocks that are distributed across multiple data storage sleds are read by a particular compute sled with least at a predefined frequency; and migrate, in response to a determination that the compute sled reads the data sub-blocks with at least the predefined frequency, the data sub-blocks to a single data storage sled.

Example 17 includes the subject matter of any of Examples 1-16, and wherein the compute engine is further to receive, from the compute sled, a selection of a key; and perform a cryptographic operation on one or more of the data sub-blocks as a function of the selection of the key.

Example 18 includes a method comprising receiving, by a network device and from a compute sled, a request to write a data block to one or more data storage sleds; determining, by the network device for each of one or more data sub-blocks within the data block and from deduplication data indicative of physical addresses of data sub-blocks, whether each data sub-block is already stored in a data storage device of a data storage sled; writing, by the network device and in the deduplication data, and in response to a determination that a data sub-block is already stored in a data storage device, a pointer to a physical address of the already-stored data sub-block in association with a logical address of the data sub-block; and writing, by the network device and to a physical address associated with a data storage device in a data storage sled, and in response to a determination that a data sub-block is not already stored in a data storage device, the data sub-block and update the deduplication data with a pointer to the physical address in association with the logical address.

Example 19 includes the subject matter of Example 18, and wherein determining whether each data sub-block is already stored in a data storage device of a data storage sled comprises generating a hash of each data sub-block; determining, from hashes associated with pointers to physical addresses present in the deduplication data, whether data has been stored in a data storage sled in association with the generated hash; and comparing, in response to a determination that data has been stored in a data storage sled in association with the generated hash, the stored data to the data sub-block to determine whether the stored data matches the data sub-block.

Example 20 includes the subject matter of any of Examples 18 and 19, and further including collecting, by the network device, telemetry data indicative of the performance and conditions of sleds connected to the network device.

Example 21 includes the subject matter of any of Examples 18-20, and wherein writing a pointer to a physical address comprises selecting a physical address as a function of the collected telemetry data.

Example 22 includes the subject matter of any of Examples 18-21, and wherein selecting a physical address as a function of the collected telemetry data comprises selecting a physical address as a function of data storage sled load data and data storage location data in the collected telemetry data.

Example 23 includes the subject matter of any of Examples 18-22, and further including determining, by the network device, whether a set of data sub-blocks that are distributed across multiple data storage sleds are accessed by a particular compute sled with least at a predefined frequency; and migrating, by the network device and in response to a determination that the compute sled accesses the data sub-blocks with at least the predefined frequency, the data sub-blocks to a single data storage sled.

Example 24 includes the subject matter of any of Examples 18-23, and further including determining, by the network device and from the collected telemetry data, whether to migrate data sub-blocks from a first storage sled to a second storage sled to balance loads across the data storage sleds.

Example 25 includes the subject matter of any of Examples 18-24, and further including migrating, by the network device, the data sub-blocks with a function to balance wear across the data storage devices in the data storage sleds.

Example 26 includes the subject matter of any of Examples 18-25, and further including migrating, by the network device, the data sub-blocks with a function to balance I/O loads across the data storage devices in the data storage sleds.

Example 27 includes the subject matter of any of Examples 18-26, and further including migrating, by the network device, the data sub-blocks from a first data storage sled to a second data storage sled with less network congestion than the first data storage sled.

Example 28 includes the subject matter of any of Examples 18-27, and further including receiving, by the network device and from the compute sled, a selection of a key; and performing, by the network device, a cryptographic operation on one or more of the data sub-blocks as a function of the selection of the key.

Example 29 includes the subject matter of any of Examples 18-28, and wherein determining whether each data sub-block is already stored in a data storage device of a data storage sled comprises performing the determination with a deduplication logic unit that is local to the network device.

Example 30 includes the subject matter of any of Examples 18-29, and wherein determining whether each data sub-block is already stored in a data storage device of a data storage sled comprises performing the determination with a deduplication logic unit that is remote from the network device.

Example 31 includes the subject matter of any of Examples 18-30, and wherein performing the determination with a deduplication logic unit that is remote from the network device comprises performing the determination with a deduplication logic unit that is located on an accelerator sled that is communicatively coupled to the network device.

Example 32 includes the subject matter of any of Examples 18-31, and further including receiving, by the network device and from the compute sled, a read request for a data block to be read; referencing, by the network device, the deduplication data to identify physical addresses of data sub-blocks within the data block to be read; accessing, by the network device, the data sub-blocks from the identified physical addresses; and providing, by the network device, the accessed data sub-blocks to the compute sled.

Example 33 includes the subject matter of any of Examples 18-32, and further including collecting, by the network device, telemetry data indicative of performance conditions of sleds connected to the network device; determining, by the network device, whether a set of data sub-blocks that are distributed across multiple data storage sleds are read by a particular compute sled with least at a predefined frequency; and migrating, by the network device and in response to a determination that the compute sled reads the data sub-blocks with at least the predefined frequency, the data sub-blocks to a single data storage sled.

Example 34 includes the subject matter of any of Examples 18-33, and further including receiving, by the network device and from the compute sled, a selection of a key; and performing, by the network device, a cryptographic operation on one or more of the data sub-blocks as a function of the selection of the key.

Example 35 includes one or more machine-readable storage media comprising a plurality of instructions stored thereon that, in response to being executed, cause a network device to perform the method of any of Examples 18-34.

Example 36 includes a network device comprising one or more processors; one or more memory devices having stored therein a plurality of instructions that, when executed by the one or more processors, cause the network device to perform the method of any of Examples 18-34.

Example 37 includes a network device comprising means for receiving, from a compute sled, a request to write a data block to one or more data storage sleds; means for determining, for each of one or more data sub-blocks within the data block and from deduplication data indicative of physical addresses of data sub-blocks, whether each data sub-block is already stored in a data storage device of a data storage sled; means for writing, in the deduplication data, and in response to a determination that a data sub-block is already stored in a data storage device, a pointer to a physical address of the already-stored data sub-block in association with a logical address of the data sub-block; and means for writing, to a physical address associated with a data storage device in a data storage sled, and in response to a determination that a data sub-block is not already stored in a data storage device, the data sub-block and update the deduplication data with a pointer to the physical address in association with the logical address.

Example 38 includes the subject matter of Example 37, and wherein the means for determining whether each data sub-block is already stored in a data storage device of a data storage sled comprises means for generating a hash of each data sub-block; means for determining, from hashes associated with pointers to physical addresses present in the deduplication data, whether data has been stored in a data storage sled in association with the generated hash; and means for comparing, in response to a determination that data has been stored in a data storage sled in association with the generated hash, the stored data to the data sub-block to determine whether the stored data matches the data sub-block.

Example 39 includes the subject matter of any of Examples 37 and 38, and further including means for collecting telemetry data indicative of the performance and conditions of sleds connected to the network device.

Example 40 includes the subject matter of any of Examples 37-39, and wherein the means for writing a pointer to a physical address comprises means for selecting a physical address as a function of the collected telemetry data.

Example 41 includes the subject matter of any of Examples 37-40, and wherein the means for selecting a physical address as a function of the collected telemetry data comprises means for selecting a physical address as a function of data storage sled load data and data storage location data in the collected telemetry data.

Example 42 includes the subject matter of any of Examples 37-41, and further including means for determining whether a set of data sub-blocks that are distributed across multiple data storage sleds are accessed by a particular compute sled with least at a predefined frequency; and means for migrating, in response to a determination that the compute sled accesses the data sub-blocks with at least the predefined frequency, the data sub-blocks to a single data storage sled.

Example 43 includes the subject matter of any of Examples 37-42, and further including means for determining, from the collected telemetry data, whether to migrate data sub-blocks from a first storage sled to a second storage sled to balance loads across the data storage sleds.

Example 44 includes the subject matter of any of Examples 37-43, and further including means for migrating the data sub-blocks with a function to balance wear across the data storage devices in the data storage sleds.

Example 45 includes the subject matter of any of Examples 37-44, and further including means for migrating the data sub-blocks with a function to balance I/O loads across the data storage devices in the data storage sleds.

Example 46 includes the subject matter of any of Examples 37-45, and further including means for migrating the data sub-blocks from a first data storage sled to a second data storage sled with less network congestion than the first data storage sled.

Example 47 includes the subject matter of any of Examples 37-46, and further including means for receiving, from the compute sled, a selection of a key; and means for performing a cryptographic operation on one or more of the data sub-blocks as a function of the selection of the key.

Example 48 includes the subject matter of any of Examples 37-47, and wherein the means for determining whether each data sub-block is already stored in a data storage device of a data storage sled comprises means for performing the determination with a deduplication logic unit that is local to the network device.

Example 49 includes the subject matter of any of Examples 37-48, and wherein the means for determining whether each data sub-block is already stored in a data storage device of a data storage sled comprises means for performing the determination with a deduplication logic unit that is remote from the network device.

Example 50 includes the subject matter of any of Examples 37-49, and wherein the means for performing the determination with a deduplication logic unit that is remote from the network device comprises means for performing the determination with a deduplication logic unit that is located on an accelerator sled that is communicatively coupled to the network device.

Example 51 includes the subject matter of any of Examples 37-50, and further including means for receiving, from the compute sled, a read request for a data block to be read; means for referencing the deduplication data to identify physical addresses of data sub-blocks within the data block to be read; means for accessing the data sub-blocks from the identified physical addresses; and means for providing the accessed data sub-blocks to the compute sled.

Example 52 includes the subject matter of any of Examples 37-51, and further including means for collecting telemetry data indicative of performance conditions of sleds connected to the network device; means for determining whether a set of data sub-blocks that are distributed across multiple data storage sleds are read by a particular compute sled with least at a predefined frequency; and means for migrating, in response to a determination that the compute sled reads the data sub-blocks with at least the predefined frequency, the data sub-blocks to a single data storage sled.

Example 53 includes the subject matter of any of Examples 37-52, and further including means for receiving, from the compute sled, a selection of a key; and means for performing a cryptographic operation on one or more of the data sub-blocks as a function of the selection of the key.

Claims

1. A network device comprising:

communication circuitry to: receive, from a compute sled, a request to write a data block to one or more data storage sleds; and

a compute engine to: determine, for each of one or more data sub-blocks within the data block and from deduplication data indicative of physical addresses of data sub-blocks, whether each data sub-block is already stored in a data storage device of a data storage sled; write, in the deduplication data and in response to a determination that a data sub-block is already stored in a data storage device, a pointer to a physical address of the already-stored data sub-block in association with a logical address of the data sub-block; and write, to a physical address associated with a data storage device in a data storage sled and in response to a determination that a data sub-block is not already stored in a data storage device, the data sub-block and update the deduplication data with a pointer to the physical address in association with the logical address.

2. The network device of claim 1, wherein to determine whether each data sub-block is already stored in a data storage device of a data storage sled comprises to:

generate a hash of each data sub-block;

determine, from hashes associated with pointers to physical addresses present in the deduplication data, whether data has been stored in a data storage sled in association with the generated hash; and

compare, in response to a determination that data has been stored in a data storage sled in association with the generated hash, the stored data to the data sub-block to determine whether the stored data matches the data sub-block.

3. The network device of claim 1, wherein the compute engine is further to collect telemetry data indicative of the performance and conditions of sleds connected to the network device.

4. The network device of claim 3, wherein to write a pointer to a physical address comprises to select a physical address as a function of the collected telemetry data.

5. The network device of claim 4, wherein to select a physical address as a function of the collected telemetry data comprises to select a physical address as a function of data storage sled load data and data storage location data in the collected telemetry data.

6. The network device of claim 3, wherein the compute engine is further to:

determine whether a set of data sub-blocks that are distributed across multiple data storage sleds are accessed by a particular compute sled with least at a predefined frequency; and

migrate, in response to a determination that the compute sled accesses the data sub-blocks with at least the predefined frequency, the data sub-blocks to a single data storage sled.

7. The network device of claim 3, wherein the compute engine is further to determine, from the collected telemetry data, whether to migrate data sub-blocks from a first storage sled to a second storage sled to balance loads across the data storage sleds.

8. The network device of claim 7, wherein the compute engine is further to migrate the data sub-blocks with a function to balance wear across the data storage devices in the data storage sleds.

9. The network device of claim 7, wherein the compute engine is further to migrate the data sub-blocks with a function to balance I/O loads across the data storage devices in the data storage sleds.

10. The network device of claim 7, wherein the compute engine is further to migrate the data sub-blocks from a first data storage sled to a second data storage sled with less network congestion than the first data storage sled.

11. The network device of claim 1, wherein the compute engine is further to:

receive, from the compute sled, a selection of a key; and

perform a cryptographic operation on one or more of the data sub-blocks as a function of the selection of the key.

12. The network device of claim 1, wherein to determine whether each data sub-block is already stored in a data storage device of a data storage sled comprises to perform the determination with a deduplication logic unit that is local to the network device.

13. One or more machine-readable storage media comprising a plurality of instructions stored thereon that, when executed by a network device cause the network device to:

receive, from a compute sled, a request to write a data block to one or more data storage sleds;

determine, for each of one or more data sub-blocks within the data block and from deduplication data indicative of physical addresses of data sub-blocks, whether each data sub-block is already stored in a data storage device of a data storage sled;

write, in the deduplication data and in response to a determination that a data sub-block is already stored in a data storage device, a pointer to a physical address of the already-stored data sub-block in association with a logical address of the data sub-block; and

write, to a physical address associated with a data storage device in a data storage sled and in response to a determination that a data sub-block is not already stored in a data storage device, the data sub-block and update the deduplication data with a pointer to the physical address in association with the logical address.

14. The one or more machine-readable storage media of claim 13, wherein to determine whether each data sub-block is already stored in a data storage device of a data storage sled comprises to:

generate a hash of each data sub-block;

determine, from hashes associated with pointers to physical addresses present in the deduplication data, whether data has been stored in a data storage sled in association with the generated hash; and

compare, in response to a determination that data has been stored in a data storage sled in association with the generated hash, the stored data to the data sub-block to determine whether the stored data matches the data sub-block.

15. The one or more machine-readable storage media of claim 13, wherein the plurality of instructions, when executed, further cause the network device to collect telemetry data indicative of the performance and conditions of sleds connected to the network device.

16. The one or more machine-readable storage media of claim 15, wherein to write a pointer to a physical address comprises to select a physical address as a function of the collected telemetry data.

17. The one or more machine-readable storage media of claim 16, wherein to select a physical address as a function of the collected telemetry data comprises to select a physical address as a function of data storage sled load data and data storage location data in the collected telemetry data.

18. The one or more machine-readable storage media of claim 15, wherein the plurality of instructions, when executed, further cause the network device to:

determine whether a set of data sub-blocks that are distributed across multiple data storage sleds are accessed by a particular compute sled with least at a predefined frequency; and

migrate, in response to a determination that the compute sled accesses the data sub-blocks with at least the predefined frequency, the data sub-blocks to a single data storage sled.

19. A method comprising:

receiving, by a network device and from a compute sled, a request to write a data block to one or more data storage sleds;

determining, by the network device for each of one or more data sub-blocks within the data block and from deduplication data indicative of physical addresses of data sub-blocks, whether each data sub-block is already stored in a data storage device of a data storage sled;

writing, by the network device and in the deduplication data, and in response to a determination that a data sub-block is already stored in a data storage device, a pointer to a physical address of the already-stored data sub-block in association with a logical address of the data sub-block; and

writing, by the network device and to a physical address associated with a data storage device in a data storage sled, and in response to a determination that a data sub-block is not already stored in a data storage device, the data sub-block and update the deduplication data with a pointer to the physical address in association with the logical address.

20. The method of claim 19, wherein determining whether each data sub-block is already stored in a data storage device of a data storage sled comprises:

generating a hash of each data sub-block;

determining, from hashes associated with pointers to physical addresses present in the deduplication data, whether data has been stored in a data storage sled in association with the generated hash; and

comparing, in response to a determination that data has been stored in a data storage sled in association with the generated hash, the stored data to the data sub-block to determine whether the stored data matches the data sub-block.