TECHNIQUES TO DETERMINE AND PROCESS METRIC DATA FOR PHYSICAL RESOURCES

Abstract

Various embodiments are generally directed to an apparatus, method and other techniques for communicating metric data between a plurality of management controllers for sleds via an out-of-band (OOB) network, the sleds comprising physical resources and the metric data to indicate one or more metrics for the physical resources. Embodiments may also include determining a physical resource of the physical resources to perform a task based at least in part on the one or more metrics, and causing the task to be performed by the physical resources.

Description

RELATED CASES

This application claims priority to U.S. Provisional Patent Application No. 62/365,969, filed Jul. 22, 2016, U.S. Provisional Patent Application No. 62/376,859, filed Aug. 18, 2016, and U.S. Provisional Patent Application No. 62/427,268, filed Nov. 29, 2016, each of which are hereby incorporated by reference in their entirety.

TECHNICAL FIELD

Embodiments described herein generally include determining and communicating metric for physical resources in a data center environment.

BACKGROUND

A computing data center may include one or more computing systems including a plurality of compute nodes that may include various compute structures (e.g., servers or sleds) and may be physically located on multiple racks. The sleds may include a number of physical resources interconnected via one or more compute structures and buses. Moreover, the sleds may be interconnected with other sleds via networking connections.

Typically, a computing data center may include a management entity to distribute workloads among the compute structures located within the racks. However, these compute structures currently fail to provide the management entity detailed system information containing performance related information, such that the management entity may make intelligent decisions when providing the workloads.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings in which like reference numerals refer to similar elements.

FIG. 1 illustrates an example of a data center.

FIG. 2 illustrates an example of a rack.

FIG. 3 illustrates an example of a data center.

FIG. 4 illustrates an example of a data center.

FIG. 5 illustrates an example of a switching infrastructure.

FIG. 6 illustrates an example of a data center.

FIG. 7 illustrates an example of a sled.

FIG. 8 illustrates an example of a data center.

FIG. 9 illustrates an example of a data center.

FIG. 10 illustrates an example of a sled.

FIG. 11 illustrates an example of a data center.

FIG. 12 illustrates an example of a data center.

FIG. 13 illustrates an example of a data center.

FIG. 14 illustrates an example of a data center.

FIG. 15 illustrates an example of a sled.

FIG. 16 illustrates an example of a data center.

FIG. 17 illustrates an example of a first logic flow diagram.

FIG. 18 illustrates an example of a second logic flow diagram.

DETAILED DESCRIPTION

Various embodiments may be generally directed to determining metric data for a number of physical resources in a data center environment and providing the metric data such that a management controller can make intelligent decisions when allocating the workloads to the physical resources. As previously mentioned, current compute structures fail to provide management controllers detailed system information containing performance metrics for the compute structures such that management controllers may make intelligent decisions when providing the workloads. Thus, embodiments discussed herein are directed to solving these other problems.

For example, embodiments discussed herein may include circuitry to determine metric data for one or more physical resources of a sled, the metric data to indicate one or more metrics for the one or more physical resources. The circuitry may also send the metric data to a pod management controller via an out-of-band (OOB) link, which may include a physical link or a virtual link, and secure. The metric data may include performance metrics and additional information such that the pod management controller may make intelligent decisions to process workloads by the physical resources. Moreover, the pod management controller may receive metric data from a plurality sleds via the OOB network, determine a physical resource of the physical resources to perform a task based at least in part on the one or more metrics, and cause the task to be performed by the physical resource. Embodiments are not limited in this manner and these other details will become apparent in the following discussion.

Reference is now made to the drawings, wherein like reference numerals are used to refer to like elements throughout. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding thereof. It may be evident, however, that the novel embodiments can be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form in order to facilitate a description thereof. The intention is to cover all modifications, equivalents, and alternatives consistent with the claimed subject matter.

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 are designed 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 twister 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 usage information for the various resources, predicts resource usage for different types of workloads based on past resource usage, 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-manipulable resource sleds. Furthermore, in the illustrative embodiment, the racks 102A, 102B, 102C, 102D include integrated power sources that receive higher current than typical for power sources. The increased current 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 204-3, and physical compute resources 205-5 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 manipulable 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 embodiments, the dual-mode switch may be a single physical network wire that may be capable of carrying Ethernet or Onmi-Path communication, which may be auto-detected by the dual-mode optical switch 515 or configured by the Pod management controller. This allows for the same network to be used for Cloud traffic (Ethernet) or High Performance Computing (HPC), typically Onmi-Path or Infiniband. Moreover, and in some instances, an Onmi-Path protocol may carry Onmi-Path communication and Ethernet communication. 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. Note that in some embodiments, the architecture may not be a leaf-spine architecture, but may be a two-ply switch architecture to connect directly to the sleds.

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. In some instances, when a sled is inserted into any given one of sled spaces 603-1 to 603-5, the corresponding MPCM may couple with a counterpart MPCM of the inserted sled. This coupling may provide the inserted sled with connectivity to both signaling infrastructure and power infrastructure of the rack in which it is housed.

Included among the types of sleds to be accommodated by rack architecture 600 may be one or more types of sleds that feature expansion capabilities. 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 919 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 “1U” server height. In such embodiments, fans 919 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 921. In various embodiments, external power source 921 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 available 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 available 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 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.

FIG. 12 illustrates an example of a data center 1200 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. 12, the data center 1200 may be similar to and include features and components previously discussed. For example, the data center 1200 may generally contain a plurality of racks 1202A to 1202D, each of which may house computing equipment including a respective set of physical resources 1205A-x to 1205D-x, where x may be any positive integer from 1 to 4. The physical resources 1205 may be contained within a number of sleds 1204A through 1204D. As mentioned, the physical resources 1205 may include resources of multiple types, such as—for example—processors, co-processors, fully-programmable gate arrays (FPGAs), memory, accelerators, and storage. In embodiments, the physical resources 1205 may include physical compute resources, physical memory resources, physical storage resources, and physical accelerator resources.

In embodiments, the physical resources 1205 may be pooled within racks and between racks. For example, physical resources 1205A-1 of sled 1204A-1 may be pooled with physical resources 1205A-3 of sled 1204A-3 to provide combined processing capabilities for workloads across sleds within the same rack, e.g. rack 1202A. Similarly, physical resources of one or more racks may be combined with physical resources of one or more other racks to create a pool of physical resources to process a workload. In one example, the physical resources 1205A-3 may be combined and pooled with physical resources of 1205B-1, which are located within rack 1202A and rack 102B, respectively. Any combination of physical resources 1205 may be pooled to process a workload and embodiments are not limited in this manner. Moreover, some embodiments may include more or less physical resources 1205, sleds 1204, and/or racks 1202 and the illustrated example should not be construed in a limiting manner.

In the illustrated example of FIG. 12, the data center 1200 may provide management functionality to monitor the physical resources 1205 and provide intelligent workload and processing capabilities. The intelligent workload capabilities may include, but are not limited to, collecting metric data for the physical resources 1205, determining processing for one or more tasks of a workload by the physical resources 1205, and causing the one or more tasks of the workload to be processed by one or more particular physical resource(s) 1205 based on the metric data and service level agreement requirements.

To perform these capabilities, embodiments include communicating low-level metric data for the physical resources 1205 to a pod management controller 1231. Moreover, the data center 1200 includes a pod management controller 1231 to provide management functionality. The pod management controller 1231 may be implemented in circuitry and logic and be part of a pod management system. The pod management controller 1231 may provide a set of application programming interfaces (API) to enable operations operating on the sleds 1204 and racks 1202 to utilize the management functionality.

In embodiments, the pod management controller 1231 may couple with one or more racks 1202 via one or more Ethernet links as part of an out-of-band (OOB) network. In one example, the OOB network may be a separate network from an optical fiber network used to communicate data between the sleds, for example. Moreover, the OOB network may be a dedicated network to communicate management and control data between the sleds 1204, racks 1202, and the pod management controller 1231. In some instances, the OOB network may support other protocols and technology to communicate metric data, such as Infiniband, Onmi-Path, and so forth. These communications may include the metric data for the physical resources 1205 for use in predicting usage of the physical resources 1205 and determining the allocation of the physical resources 1205 for processing current and future workloads. These and other details will become more apparent with the following description.

FIG. 13 illustrates an example of a data center management architecture 1300 that may 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. The data center management architecture 1300 of FIG. 13 illustrates a rack 1302 having a number of sleds 1304-1 through 1304-n, where n may be any positive integer. Note that FIG. 13 only illustrates a single rack 1302 having sleds 130-n coupled with a pod management system 1333. However, embodiments are not limited in this manner, as previously discussed above in FIG. 12, for example.

Each of the sleds 1304-n includes a number of components, including physical resources 1305, a management controller 1362, and Ethernet (ETH) circuitry 1352. As will be discussed in more detail with respect to FIG. 15, each sled 1304 may also include an MPCM having an ETH connector to couple with a corresponding ETH connector to enable communication via the Ethernet links.

In embodiments, the ETH circuitry 1352 may enable communication via one or more Ethernet links of an OOB network. In some embodiments, the ETH circuitry 1352 may enable gigabyte communications with other devices, such as a pod management system 1333. The ETH circuitry 1352 may be a media-independent interface, which may use any network signal transmission media. The media-independent interface may be a reduced media-independent interface (RMII), gigabit media-independent interface (GMII), reduced gigabit media-independent interface (RGMII), 10-gigabit media-independent interface (XGMII) and serial gigabit media-independent interface (SGMII), for example.

In embodiments, the ETH circuitry 1352 may support communications via one or more architectures and data structures. For example, the ETH circuitry 1352 is capable of communicating data utilizing a representational state transfer (REST) architecture and in a JavaScript Object Notation (JSON) data format. In one example, the ETH circuitry 1352 may support a set of APIs and schema, such as Redfish®, to enable communication of data between the sleds 1304, the racks 1302, and the pod management controller 1331. In embodiments, the ETH circuitry 1352 may be coupled with and include memory to store the functions required to operate in accordance with the REST architecture and communicate data in JSON data format. For example, the ETH circuitry 1352 may be coupled with a 256 megabytes (MBs) of error correction control (ECC) memory to run an embedded operation system, e.g. Linux, to provide the ETH interface (Redfish®). Embodiments are not limited in this manner and other web services architectures may be utilized to communicate the metric data and be consistent with embodiments discussed herein.

In embodiments, each of the sleds 1304-n may also include a management controller 1362-n to collect and determine metric data for the physical resources 1305-n. A management controller 1362 also provides management functionality including sending the metric data in a data structure to a pod management controller 1331. In some instances, the management controller 1362 may be part of an Intelligent Platform Management Interface (IPMI) architecture and may be a baseboard management controller (BMC) or specialized service processor that monitors the physical state and operational state of the physical resources 1305 using sensors and communicating with the physical resources 1305 themselves. In some instances, the management controller 1362 may be a sled management controller. Embodiments are not limited in this manner. For example, in some embodiments, the metric data may also be communicated to the pod management controller 1331 as new original equipment manufacturer (OEM) records in the system management basic input/output (SMBIOS) records.

The management controller 1362 may collect metric data such as temperature, humidity, power-supply voltage, fan speeds, communication parameters and operating system functions. If metrics associated with these variables are determined to be out of specification or do not meet one or more service level agreement requirements, the management controller 1362 may notify or send the metric data to the pod management controller 1331. Moreover, and in some embodiments, the metric data may be reported or sent to the pod management controller 1331 on a periodic or semi-periodic basis, even when the variables are not out of specification. Embodiments are not limited in this manner.

In some embodiments, the management controller 1362 may collect metric data specific to a type of physical resource 1305. For example, a physical resource 1305 may be a physical memory resource and the management controller 1362 may collect metric data, such as a number of memory channels, a memory bandwidth, a memory size, a memory type, read/write parameters, a memory speed, read/write latency, partitioning information, high bandwidth memory size, socket interconnect latency, interleaving or non-interleaving indication, whether two-level memory (2LM) is utilized, near and far memory type of 2LM, size of 2LM, performance of 2LM region and so forth. Note that embodiments may also include multi-level memory, e.g. 3LM, wherein near memory (first layer) may be connected via a physical interconnect on the same baseboard and far memory may be connected via a fiber interconnect and be in another separate sled. The separate may include volatile memory (second layer), and 3D XPoint memory (third layer), for example. some embodiments, the management controller 1362 may provide advanced configuration and power interface (ACPI) static resource affinity table (SRAT) information and hierarchical memory attribute table (HMAT) information. via the OOB network. Typically, these low-level metrics associated with physical memory resources is not provided to the pod management controller 1331. Thus, by providing this low-level metric data to the pod management controller 1331, the pod management controller 1331 may make more intelligent decisions when placing tasks for workloads for processing.

In another example, the management controller 1362 may provide metric data for physical compute resources or processor, such as a processor identifier, a processor cache capability, a processor topology, a processor cache topology, processor-to-processor latency, bandwidth information, performance (speed/throughput) metrics, and so forth. In a third example, a physical resource 1305 may be a physical storage resource and the management controller 1362 may collect metric data such as a storage throughput, a storage input/output operations per second (IOPS) metric, a storage latency, a storage size, and a storage utilization. Embodiments are not limited to these examples. For instance, a physical resource 1305 may be an accelerator resource and accelerator-related metrics may be determined for the accelerator, such as a card link width associated with a Peripheral Component Interconnect Express (PCIe) card, a solid-state drive (SSD), a field programmable gate array (FPGA) card, and so forth.

In embodiments, the management controller 1362 may collect and determine the metric data for each of the physical resources 1305 utilizing any number of methods or techniques. For example, the management controller 1362 may include sensors to detect metric data itself. In another example, the management controller 1362 may determine the metric data based on sensors and determinations made by the physical resources 1305 themselves. More specifically, the physical resources 1305 may be able to monitor such metrics as throughput, usage, processing time, read/write speeds, etc. and communicate the metric data to the management controller 1362. The physical resources 1305 may also store metric data about itself within a memory location, fuse logic, or so forth that may be polled by the management controller 1362 to collect the metric data. For example, memory resource may store an indication of memory size, memory type, memory channels, memory bandwidth, etc. A processor resource may store similar information, such as processor speed, processor topology, processor type, and so forth. Similar metrics may also be polled by a management controller 1362 with respect to storage resources, such as storage size, storage throughput (read/write times), storage type, and so forth. Embodiments are not limited in this manner and the management controller 1362 may collect metric data via other means including receiving or retrieving information from an operating system.

The management controller 1362 may collect and determine the metric data on a periodic or semi-periodic basis. In some instances, the metric data may be collected on a periodic/semi-periodic basis based on a user setting, a factory (time of manufacturer) setting, service level agreement requirements, and so forth. Embodiments are not limited in this manner.

The management controller 1362 may be capable of communicating via one or more different interface or bus types to collect the metric data. For example, the management controller 1362 may be coupled with the one or more physical resources via a low pin count (LPC) bus, a system management bus (SMBus), an Inter-Integrated (I2C) bus, an IPMI utilizing the SMBus, and a serial port. These interfaces and buses may be used to collect and determine metric data from the different physical resources.

The management controller 1362 may also be coupled with the ETH circuitry 1352 via one or more buses/interfaces to communicate the metric data with the pod management controller 1331. More specifically, the management controller 1362 may utilize the ETH circuitry 1352 to communicate the metric data via a REST architecture and in a JSON data format, for example.

FIG. 14 illustrates an example of a data center management architecture 1400 that may 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. The data center management architecture 1400 of FIG. 14 may be similar to the data center management architecture 1300 illustrated in FIG. 13; however, the rack 1402 may include a rack management controller 1464 to communicate the metric data with the sleds 1404 and the pod management controller 1431. For example, the rack management controller 1464 may collect (or receive) the metric data from the sleds 1404 and send the metric data to the pod management system 1433 and pod management controller 1431. In a data center management architecture 1400, each rack, such as rack 1402, may include a rack management controller 1464 to collect metric data and other information from each of the sleds 1404.

In embodiments, the rack management controller 1464 may include instructions and logic, such as Intel's® Pooled System Management Engine (PSME), to collect, manage, and communicate the metric data for each of the sleds 1404 in the rack 1402. The rack management controller 1464 may also include ETH circuitry 1456, which may be similar to the ETH circuitry 1452 of the sleds 1404. For example, the ETH circuitry 1456 may be a RGMII interface and capable of communicating data utilizing a REST Architecture in a JSON data structure, such as Redfish®. In some instances, the rack management controller 1464 may operate as a server including a REST API to gather the metric data from the sleds 1405 and present/send the metric data to the pod management controller 1431. The sleds 1405, rack management controller 1464, and the pod management controller 1431 may communicate the metric data through a JSON-RPC as a transport and JSON as a data structure, for example. Moreover and as similarly discussed above, these communications may be made in an OOB network environment to prevent interference and bandwidth usage with other data. Embodiments are not limited in this manner.

FIG. 15 illustrates an example of a sled 1504 that may be representative of a sled designed for use in conjunction with the racks discussed herein, for example. In embodiments, sled 1504 may be similar to and have similar components and functionality as sled 1004 discussed in FIG. 15. Sled 1504 may feature an MPCM 1516 that which may include an optical connector 1516A, a power connector 1516B, and an ETH connector 1516C, and that is designed to couple with a counterpart MPCM of a sled space in conjunction with insertion of MPCM 1516 into that sled space. Coupling MPCM 1516 with such a counterpart MPCM may cause power connector 1516B to couple with a power connector comprised in the counterpart MPCM. This may enable physical resources 1505 of sled 1504 to source power from an external source, via power connector 1516B and power transmission media 1524 that conductively couples power connector 1516 to physical resources 1505.

Sled 1504 may also include dual-mode optical network interface circuitry 1526. Dual-mode optical network interface circuitry 1526 may include 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, as previously discussed in FIGS. 9 and 10. In some embodiments, dual-mode optical network interface circuitry 1526 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 1526 may include one or more optical transceiver modules 1527, 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 1516 with a counterpart MPCM of a sled space in a given rack may cause optical connector 1516A to couple with an optical connector comprised in the counterpart MPCM. This may establish optical connectivity between optical cabling of the sled and dual-mode optical network interface circuitry 1526, via each of a set of optical channels 1525. Dual-mode optical network interface circuitry 1526 may communicate with the physical resources 1505 of sled 1504 via electrical signaling media 1528.

The sled 1504 may also include a management controller 1562, which may be the same as or similar to management controller 1362 of FIG. 13 and management controller 1462 of FIG. 14. The management controller 1562 may determine and collect metric data for physical resources 1505, including but not limited to, physical memory resources 1505-1, physical compute resources 1505-2, physical storage resources 1505-3, and physical accelerator resources 1505-4. Embodiments are not limited in this manner.

A physical memory resource 1505-1 may be any type of memory, such as any machine-readable or computer-readable media capable of storing data, including both volatile and non-volatile memory. In some embodiments, the machine-readable or computer-readable medium may include a non-transitory medium. Moreover, a physical memory resource 1505-1 may include one or more higher speed memory units, such as read-only memory (ROM), random-access memory (RAM), dynamic RAM (DRAM), Double-Data-Rate DRAM (DDRAM), synchronous DRAM (SDRAM), static RAM (SRAM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, polymer memory such as ferroelectric polymer memory, ovonic memory, phase change or ferroelectric memory, silicon-oxide-nitride-oxide-silicon (SONOS) memory, magnetic or optical cards, an array of devices such as Redundant Array of Independent Disks (RAID) drives, solid state memory devices (e.g., USB memory, solid state drives (SSD), 3D Xpoint®, and any other type of storage media suitable for storing information. Embodiments are not limited to these examples.

A physical compute resource 1505-2 may be any type of circuitry capable of processing information. Moreover, a physical compute resources 1505-2 may be implemented using any processor or logic device. The physical compute resource 1505-2 may be one or more of any type of computational element, such as but not limited to, a microprocessor, a processor, central processing unit, digital signal processing unit, dual core processor, mobile device processor, desktop processor, single core processor, a system-on-chip (SoC) device, complex instruction set computing (CISC) microprocessor, a reduced instruction set (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, or any other type of processor or processing circuit on a single chip or integrated circuit. The physical compute resource 1505-2 may be connected to and communicate with the other physical resources 1505 of the computing system via an interconnect, such as one or more buses, control lines, and data lines.

In embodiments, a physical storage resource 1505-3 may be any type of storage, and may be implemented as a non-volatile storage device such as, but not limited to, a magnetic disk drive, optical disk drive, tape drive, an internal storage device, an attached storage device, flash memory, battery backed-up SDRAM (synchronous DRAM), and/or a network accessible storage device. In embodiments, a physical storage resource 1505-3 may include technology to increase the storage performance enhanced protection for valuable digital media when multiple hard drives are included, for example. Further examples of physical storage resource 1505-3 may include a hard disk, floppy disk, Compact Disk Read Only Memory (CD-ROM), Compact Disk Recordable (CD-R), Compact Disk Rewriteable (CD-RW), optical disk, magnetic media, magneto-optical media, removable memory cards or disks, various types of DVD devices, a tape device, a cassette device, or the like. The embodiments are not limited in this context.

A physical accelerator resource 1505-4 may be any type of accelerator device designed to increase processing power of a processor, such as the physical compute resource 1505-2. The physical accelerator resource 1505-4 accelerates transmission or processing beyond processor capabilities. In one example, a physical accelerator resource 1505-4 may compute faster floating-point units (FPUs) by assisting in math calculations or by increasing speed. In another example, the physical accelerator resource 1505-4 may be a graphics processing units (GPUs) for 3-D images or faster graphic displays. Embodiments, the physical accelerator resource 1505-4 may be implemented as field programmable gate arrays (FPGAs); however, embodiments are not limited in this manner.

The management controller 1562 may collect metric data for one or more of the physical resources 1505 via one or more interconnects 1538 and electrical signals. The interconnects 1538 may be a low pin count (LPC) bus, a system management bus (SMBus), an Inter-Integrated (I2C) bus, an IPMI utilizing the SMBus, and a serial port. Embodiments are not limited to these examples.

In embodiments, the management controller 1562 may communicate the metric data to a pod management controller. In some instances, the management controller 1562 may communicate the metric data to the pod management controller via a rack management controller. To send the metric data, the management controller 1562 may utilize the ETH circuitry 1552 to send the metric data via a REST architecture and in a JSON data format, as previously discussed. The ETH circuitry 1552 may provide a layered architecture to communicate the metric in the REST architecture and a JSON data format, for example. The management controller 1562 may send the metric data via one or more interconnects 1538 as electrical signals, for example.

The ETH circuitry 1552 may process the metric data to communicate it via the REST architecture in a JSON data format. For example, the ETH circuitry 1552 may put the metric data in one or more packets having a JSON data format. The ETH circuitry 1552 may send the metric data via the ETH connector 1515C coupled with a counterpart ETH connector in a sled space in conjunction with insertion of MPCM 1515 into that sled space. In some instances, the ETH connector 1515C may be a modular connector or uniquely designed connector to couple with the counterpart ETH connector in the sled space. In one example, the ETH connector 1515C may have the same wiring pinout as a registered jack 45 (RJ45) modular connector. However, the ETH connector 1515C may be designed differently than a standard RJ45 connector such that it may couple with the counterpart ETH connector in the sled space.

In some embodiments, the management controller 1562 and ETH circuitry 1552 may perform a serial-to-local area network (LAN) conversion to communicate the metric data to the pod management controller. For example, the management controller 1562 may collect the metric data via one or more serial links, convert the data for transmission into packets in a JSON data format, and communicate the metric data to the pod management controller. Embodiments are not limited to these examples.

FIG. 16 illustrates an example of a data center system 1600 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. 16, the data center system 1600 may be similar to and include features and components previously discussed. For example, the data center system 1600 may generally contain a plurality of racks 1602A to 1602D, each of which may house computing equipment including a respective set of physical resources 1605A-x to 1605D-x, where x may be any positive integer from 1 to 4. The physical resources 1605 may be contained within a number of sleds 1604A through 1604D. As mentioned, the physical resources 1605 may include resources of multiple types, such as—for example—processors, co-processors, fully-programmable gate arrays (FPGAs), memory, accelerators, and storage.

In embodiments, the sleds 1604 may communicate metric data to a pod management controller 1631, as previously discussed. For example, the sleds 1604 may each include a management controller (not shown) that may collect the metric data of the physical resources 1605 and send the metric data to the pod management controller 1631 either directly or via a rack management controller (not shown). Moreover, the metric data may be sent to the pod management controller 1631 via an OOB network on one or more ETH links utilizing a REST architecture and in a JSON data format.

The pod management controller 1631 utilize the metric data to determine physical resources 1605 for processing one or more tasks of a workload. For example, the pod management controller 1631 may implement an orchestration layer, such as OpenStack®, to consume the metric data to allocate physical resources for processing workloads. In embodiments, the pod management controller 1631 may utilize the metric data in combination with service level agreement (SLA) requirements to cause tasks for workloads to be processed by physical resources while maintaining the requirements stipulated in a SLA. The SLA may be based on a policy-based storage management system to help evaluate and maintain an adequate level performance for a data center. The SLA may specify a set of one or more values or metrics relating to one or more specific, measurable performance characteristics and specifying one or more desired or required levels of service to be provided to a workload including one or more tasks. Some requirements may include, latency, cost, protection against local failures or corruption, geographic dispersion, efficiency, throughput, processing times, etc. Thus, SLA requirements can be defined regarding any one or more of these characteristics, and other characteristics. By collecting the metric data and determine actual performance relative to SLA, it can be determined whether a data center is performing adequately, and adjustments to the state of the data center system can be made if it is not. For example, the pod management controller 1631 may adjust, send, cause, etc. which physical resources are processing particular tasks of workloads to ensure that the requirements of the SLA are being met. More specifically, processing cycles on physical compute resources, memory read/writes of physical memory resources, data storage of physical storage resources, and processing cycles of accelerators may be allocated to workloads based on the SLA requirements and metric data.

In embodiments, the pod management controller 1631 may determine SLA requirements from data stored in a memory or storage, such as data store 1677. The SLA requirements may be stored in the data store 1677 based on user input or computer determinations specifying particular SLA requirements for workloads. Thus, a pod management controller 1631 may receive an indication of a workload to be processed by the data center 1600 from one or more clients 1679. The pod management controller 1631 can determine the SLA requirements for the workload based on the data in the data store 1677. For example, the pod management controller 1631 may perform a lookup and retrieve the SLA requirements for the workload based on an identifier identifying the workload.

The pod management controller 1631 may utilize the SLA requirements for the workload and the metric data received from the racks 1602 to determine which physical resources 1605 are to process one or more tasks of the workload. For example, the pod management controller 1631 may determine which physical resources 1605 are capable to process one or more tasks of a workload to meet SLA requirements for the workload. The pod management controller 1631 may cause the one or more tasks to be processed by the determined physical resources 1605. Note that the metric data may be communicated between the racks 1602 and the pod management controller 1631 via the OOB network; however, the one or more tasks may be communicated to the racks 1602 and particular resources 1605 via a different network, such as an optical fiber network. Embodiments are not limited in this manner.

FIG. 17 illustrates an embodiment of logic flow 1700. The logic flow 1700 may be representative of some or all of the operations executed by one or more embodiments described herein. For example, the logic flow 1700 may illustrate operations performed by a pod management controller, as discussed herein. However, embodiments are not limited in this, and one or more operations may be performed by other components or systems discussed herein.

At block 1702, the logic flow 1700 includes to determining metric data for one or more physical resources. As previously discussed, a pod management controller may receive metric data from a management controller of a sled and a rack management controller of a rack having one or more sleds. The metric data may be collected and determined by a management controller of a sled having physical resources and provided via an OOB network.

At block 1704, the logic flow 1700 includes determining one or more tasks of workload that are to be processed by a data center. In some instances, a pod management controller may receive the tasks and workload or an indication of the task and workload. The indication may identify tasks and workload, for example. The tasks may include any type of operations, jobs, and processing that may be completed by the physical resources. For example, a task includes instructions to be processed by a physical compute resource, read/write requests for a physical memory resource, read/write requests for a physical storage resource and instructions to be processed by a physical accelerator resource.

At block 1706, the logic flow 1700 includes determining one or more SLA requirements for the workload. For example, a pod management controller may retrieve SLA requirement data from data associated with the workload. The SLA requirements may specify one or more requirements for the workload and the tasks, such as processing, throughput, IOPS, read/write speeds, etc. Embodiments are not limited in this manner.

At block 1708, the logic flow 1700 may determine one or more physical resources to process one or more tasks for a workload. For example, a pod management controller may determine which physical resources are capable of processing the tasks while meeting the SLA requirements for the tasks. Note that the pod management controller may utilize a single physical resource to perform a task or pool two more physical resources to process the task. In embodiments, the pod management controller may determine which physical resources based on the metric data indicating which of the physical resources can meet the SLA requirements.

At block 1710, the logic flow 1700 includes causing a task to be performed by the one or more physical resources determine to be capable of meeting the SLA requirement for the task. For example, a pod management controller may communicate information to one or more clients or other systems indicating which physical resources are to perform/process one or more task of a workload. Embodiments are not limited in this manner.

Although logic flow 1700 illustrates particular operations occurring in a particular order, embodiments are not limited in this manner and some operations may occur before, after or during other operations. Also, logic flow 1700 may repeat any number of times and embodiments are not limited in this manner.

FIG. 18 illustrates an embodiment of logic flow 1800. The logic flow 1800 may be representative of some or all of the operations executed by one or more embodiments described herein. For example, the logic flow 1800 may illustrate operations performed by a management controller of a sled, as discussed herein. However, embodiments are not limited in this, and one or more operations may be performed by other components or systems discussed herein.

At block 1802, the logic flow 1800 includes determining metric data for one or more physical resources of a sled. For example, a management controller may determine and collect metric data for one or more physical resources including, but not limited to, a physical memory resource, a physical compute resource, a physical storage resource, and a physical accelerator resource. The management controller may collect the metric data via one or more sensors, from the physical resources, and from an operating system for a sled. Embodiments are not limited in this manner.

At block 1804, the logic flow 1800 includes sending the metric data to a pod management controller. In some instances, a management controller may send the metric data directly to the pod management controller in a REST architecture in a JSON data format via an OOB network. In another example, the management controller may send the metric data to the pod management controller via a rack management controller. A rack management controller may receive metric data from any number of sleds and physical resources of sleds to send to the pod management controller. The rack management controller may also send the metric in a REST architecture and JSON data format. Note that in both examples, the metric data may be sent to the pod management controller via an OOB network such that the metric data does not interfere with processing and data transfer on other networks, such as an optical fiber network.

At block 1806, the logic flow 1800 may include receiving a task to be processed by one or more physical resources of a sled. In some instances, the task may be part of workload being processed by the data center and may be sent to the sled having the one or more physical resources based on the metric data. Embodiments are not limited in this manner.

The detailed disclosure now turns to providing examples that pertain to further embodiments. Examples one through twenty-five (1-25) provided below are intended to be exemplary and non-limiting.

In a first example, a system, a device, an apparatus, and so forth may include a pod management controller receive metric data from a plurality of management controllers for sleds via an out-of-band (OOB) network, the sleds comprising physical resources and the metric data to indicate one or more metrics for the physical resources, determine a physical resource of the physical resources to perform a task based at least in part on the one or more metrics, and cause the task to be performed by the physical resource.

In a second example and in furtherance of the first example, a system, a device, an apparatus, and so forth including the pod management controller to determine the physical resource to perform the task based on the metric data indicating the physical resource is capable of meeting a requirement of a service level agreement associated with the task.

In a third example and in furtherance of any of the previous examples, a system, a device, an apparatus, and so forth including the pod management controller to receive the metric data from the plurality of management controllers for the sleds located within a single rack.

In a fourth example and in furtherance of any of the previous examples, a system, a device, an apparatus, and so forth including the logic to receive the metric data from the plurality of management controllers for the sleds located within two or more racks.

In a fifth example and in furtherance of any of the previous examples, a system, a device, an apparatus, and so forth including the pod management controller to receive the metric data via the OOB network utilizing a representational state transfer (REST) architecture and in a JavaScript Object Notation (JSON) data format.

In a sixth example and in furtherance of any of the previous examples, a system, a device, an apparatus, and so forth including the pod management controller to cause processing of the physical resources comprising one or more physical memory resource and the metric data for each of the physical memory resources comprising the physical resources comprising one or more physical memory resource and the metric data for each of the physical memory resources comprising one or more of an indication whether the physical memory resources are interleaved or non-interleaved, one or more of a memory throughput, a memory input/output operations per second (IOPS) metric, a memory latency, a memory size, and a memory utilization.

In a seventh example and in furtherance of any of the previous examples, a system, a device, an apparatus, and so forth including the pod management controller to cause processing of the physical resources comprising one or more physical compute resource and the metric data for each of the physical compute resources comprising one or more of a processor identifier, a processor cache capability, a processor topology, a processor cache topology, processor-to-processor link access latency, and processor bandwidth information.

In an eighth example and in furtherance of any of the previous examples, a system, a device, an apparatus, and so forth including the pod management controller to cause processing of the physical resources comprising one or more physical storage resources and the metric data for each of the one or more physical storage resource comprising one or more of a storage throughput, a storage input/output operations per second (IOPS) metric, a storage latency, a storage size, and a storage utilization.

In a ninth example and in furtherance of any of the previous examples, a system, a device, an apparatus, and so forth including the pod management controller to receive the metric data via a rack management controller receiving metric data from a plurality of sleds of one or more racks.

In a tenth example and in furtherance of any of the previous examples, embodiments may include a non-transitory computer-readable storage medium, comprising a plurality of instructions, that when executed, enable processing circuitry to receive metric data from a plurality of management controllers for sleds via an out-of-band (OOB) network, the sleds comprising physical resources and the metric data to indicate one or more metrics for the physical resources, determine a physical resource of the physical resources to perform a task based at least in part on the one or more metrics, and cause the task to be performed by the physical resource.

In an eleventh example and in furtherance of any of the previous examples, embodiments may include a non-transitory computer-readable storage medium, comprising a plurality of instructions, that when executed, enable processing circuitry to determine the physical resource to perform the task based on the metric data indicating the physical resource is capable of meeting a requirement of a service level agreement associated with the task.

In a twelfth example and in furtherance of any of the previous examples, embodiments may include a non-transitory computer-readable storage medium, comprising a plurality of instructions, that when executed, enable processing circuitry to receive the metric data from the plurality of management controllers for the sleds located within a single rack.

In a thirteenth example and in furtherance of any of the previous examples, embodiments may include a non-transitory computer-readable storage medium, comprising a plurality of instructions, that when executed, enable processing circuitry to receive the metric data from the plurality of management controllers for the sleds located within two or more racks.

In a fourteenth example and in furtherance of any of the previous examples, embodiments may include a non-transitory computer-readable storage medium, comprising a plurality of instructions, that when executed, enable processing circuitry to receive the metric data via the OOB network utilizing a representational state transfer (REST) architecture and in a JavaScript Object Notation (JSON) data format.

In a fifteenth example and in furtherance of any of the previous examples, embodiments may include a non-transitory computer-readable storage medium, comprising a plurality of instructions, that when executed, enable processing circuitry to receive the metric data via a rack management controller receiving metric data from a plurality of sleds of one or more racks.

In a sixteenth example and in furtherance of any of the previous examples, a system, a device, an apparatus, and so forth including a management controller to determine metric data for one or more physical resources of a sled, the metric data to indicate one or more metrics for the one or more physical resources, send the metric data to a pod management controller via an out-of-band (OOB) link, receive a task for processing by one or more of the physical resources.

In a seventeenth example and in furtherance of any of the previous examples, a system, a device, an apparatus, and so forth including the management controller to send the metric data via the OOB link utilizing a representational state transfer (REST) architecture and in a JavaScript Object Notation (JSON) data format.

In an eighteenth example and in furtherance of any of the previous examples, a system, a device, an apparatus, and so forth including the management controller to send the metric data to the pod management controller via a rack management controller.

In a nineteenth example and in furtherance of any of the previous examples, embodiments may include a non-transitory computer-readable storage medium, comprising a plurality of instructions, that when executed, enable processing circuitry to determine metric data for one or more physical resources of a sled, the metric data to indicate one or more metrics for the one or more physical resources, and send the metric data to a pod management controller via an out-of-band (OOB) link.

In a twentieth example and in furtherance of any of the previous examples, embodiments may include a non-transitory computer-readable storage medium, comprising a plurality of instructions, that when executed, enable processing circuitry to send the metric data via the OOB link utilizing a representational state transfer (REST) architecture and in a JavaScript Object Notation (JSON) data format.

In a twenty-first example and in furtherance of any of the previous examples, embodiments may include a non-transitory computer-readable storage medium, comprising a plurality of instructions, that when executed, enable processing circuitry to send the metric data to the pod management controller via a rack management controller.

In a twenty-second example and in furtherance of any of the previous examples, embodiments may include one or more methods to perform any combination of the above-recited examples or other methods/logic flows discussed herein.

Some embodiments may be described using the expression “one embodiment” or “an embodiment” along with their derivatives. These terms mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. The appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment. Further, some embodiments may be described using the expression “coupled” and “connected” along with their derivatives. These terms are not necessarily intended as synonyms for each other. For example, some embodiments may be described using the terms “connected” and “coupled” to indicate that two or more elements are in direct physical or electrical contact with each other. The term “coupled,” however, may also mean that two or more elements are not in direct contact with each other, but yet still co-operate or interact with each other.

It is emphasized that the Abstract of the Disclosure is provided to allow a reader to quickly ascertain the nature of the technical disclosure. It is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims. Also, in the preceding Detailed Description, it can be seen that various features are grouped together in a single embodiment for the purpose of streamlining the disclosure. This method of disclosure is not to be interpreted as reflecting an intention that the claimed embodiments require more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive subject matter lies in less than all features of a single disclosed embodiment. Thus the following claims are at this moment incorporated into the Detailed Description, with each claim standing on its own as a separate embodiment. In the appended claims, the terms “including” and “in which” are used as the plain-English equivalents of the respective terms “comprising” and “wherein,” respectively. Moreover, the terms “first,” “second,” “third,” and so forth, are used merely as labels, and are not intended to impose numerical requirements on their objects.

What has been described above includes examples of the disclosed architecture. It is, of course, not possible to describe every conceivable combination of components and/or methodologies, but one of ordinary skill in the art may recognize that many further combinations and permutations are possible. Accordingly, the novel architecture is intended to embrace all such alterations, modifications and variations that fall within the spirit and scope of the appended claims.

Claims

1. A system, comprising:

a pod management controller coupled with sleds via an out-of-band (OOB) network, the pod management controller to: receive metric data from a plurality of management controllers for the sleds via the OOB network, the sleds comprising physical resources and the metric data to indicate one or more metrics for the physical resources; determine a physical resource of the physical resources to perform a task based on the one or more metrics; and send the task to be performed by the physical resource of one of the sleds.

2. The system of claim 1, the pod management controller to determine the physical resource to perform the task based on the metric data indicating the physical resource is capable of meeting a requirement of a service level agreement associated with the task.

3. The system of claim 1, the logic to receive the metric data from the plurality of management controllers for the sleds located within a single rack.

4. The system of claim 1, the logic to receive the metric data from the plurality of management controllers for the sleds located within two or more racks.

5. The system of claim 1, the logic to receive the metric data via the OOB network utilizing a representational state transfer (REST) architecture and in a JavaScript Object Notation (JSON) data format.

6. The system of claim 1, the physical resources comprising one or more physical memory resource and the metric data for each of the physical memory resources comprising one or more of an indication whether physical memory resources are interleaved or non-interleaved, one or more of a memory throughput, a memory input/output operations per second (IOPS) metric, a memory latency, a memory size, and a memory utilization.

7. The system of claim 1, the physical resources comprising one or more physical compute resource and the metric data for each of the physical compute resources comprising one or more of a processor identifier, a processor cache capability, a processor topology, a processor cache topology, processor-to-processor link access latency, and processor bandwidth information.

8. The system of claim 1, the physical resources comprising one or more physical storage resources and the metric data for each of the one or more physical storage resource comprising one or more of a storage throughput, a storage input/output operations per second (IOPS) metric, a storage latency, a storage size, and a storage utilization.

9. The system of claim 1, the pod management controller to receive the metric data via a rack management controller receiving metric data from the sleds of one or more racks.

10. A non-transitory computer-readable storage medium, comprising a plurality of instructions, that when executed, enable processing circuitry to:

receive, by a pod management controller, metric data from a plurality of management controllers for sleds via an out-of-band (OOB) network, the sleds comprising physical resources and the metric data to indicate one or more metrics for the physical resources;

determine, by the pod management controller, a physical resource of the physical resources to perform a task based at least in part on the one or more metrics; and

send, by the pod management controller, the task to be performed by the physical resource of one of the sleds.

11. The computer-readable storage medium of claim 10, comprising a plurality of instructions, that when executed, enable processing circuitry to determine the physical resource to perform the task based on the metric data indicating the physical resource is capable of meeting a requirement of a service level agreement associated with the task.

12. The computer-readable storage medium of claim 10, comprising a plurality of instructions, that when executed, enable processing circuitry to receive the metric data from the plurality of management controllers for the sleds located within a single rack.

13. The computer-readable storage medium of claim 10, comprising a plurality of instructions, that when executed, enable processing circuitry to receive the metric data from the plurality of management controllers for the sleds located within two or more racks.

14. The computer-readable storage medium of claim 10, comprising a plurality of instructions, that when executed, enable processing circuitry to receive the metric data via the OOB network utilizing a representational state transfer (REST) architecture and in a JavaScript Object Notation (JSON) data format.

15. The computer-readable storage medium of claim 10, the physical resources comprising one or more physical memory resource and the metric data for each of the physical memory resources comprising one or more of an indication whether physical memory resources are interleaved or non-interleaved, one or more of a memory throughput, a memory input/output operations per second (IOPS) metric, a memory latency, a memory size, and a memory utilization.

16. The computer-readable storage medium of claim 10, the physical resources comprising one or more physical compute resource and the metric data for each of the physical compute resources comprising one or more of a processor identifier, a processor cache capability, a processor topology, a processor cache topology, processor-to-processor link access latency, and processor bandwidth information.

17. The computer-readable storage medium of claim 10, the physical resources comprising one or more physical storage resources and the metric data for each of the one or more physical storage resource comprising one or more of a storage throughput, a storage input/output operations per second (IOPS) metric, a storage latency, a storage size, and a storage utilization.

18. The computer-readable storage medium of claim 10, comprising a plurality of instructions, that when executed, enable processing circuitry to receive the metric data via a rack management controller receiving metric data from a plurality of sleds of one or more racks.

19. An apparatus, comprising:

a management controller of a sled coupled with a pod management controller via an out-of-band (OOB) link, the management controller to: determine metric data for one or more physical resources of the sled, the metric data to indicate one or more metrics for the one or more physical resources; send the metric data to a pod management controller via the OOB link; and receive a task to be processed by at least one of the one or more physical resources of the sled.

20. The apparatus of claim 19, the logic to send the metric data via the OOB link utilizing a representational state transfer (REST) architecture and in a JavaScript Object Notation (JSON) data format.

21. The apparatus of claim 19, the physical resources comprising one or more physical memory resource and the metric data for each of the physical memory resources comprising one or more of an indication whether physical memory resources are interleaved or non-interleaved, one or more of a memory throughput, a memory input/output operations per second (IOPS) metric, a memory latency, a memory size, and a memory utilization.

22. The apparatus of claim 19, the physical resources comprising one or more physical compute resource and the metric data for each of the physical compute resources comprising one or more of a processor identifier, a processor cache capability, a processor topology, a processor cache topology, processor-to-processor link access latency, and processor bandwidth information.

23. The apparatus of claim 19, the physical resources comprising one or more physical storage resources and the metric data for each of the one or more physical storage resources comprising one or more of a storage throughput, a storage input/output operations per second (IOPS) metric, a storage latency, a storage size, and a storage utilization.

24. The apparatus of claim 19, the management controller to send the metric data to the pod management controller via a rack management controller.

25. The apparatus of claim 19, the physical resources comprising one or more of a physical memory resource, a physical compute resource, a physical storage resource, and a physical accelerator resource.