TECHNOLOGIES FOR AUTOMATED NETWORK CONGESTION MANAGEMENT
Technologies for congestion management include multiple storage sleds, compute sleds, and other computing devices in communication with a resource manager server. The resource manager server discovers the topology of the sleds and one or more layers of network switches that connect the sleds. The resource manager server constructs a model of network connectivity between the sleds and the switches based on the topology, and determines an oversubscription of the network based on the model. The oversubscription is based on available bandwidth for the layer of switches and maximum potential bandwidth used by the sleds. The resource manager server determines bandwidth limits for each sled and programs each sled with the corresponding bandwidth limit. Each sled enforces the programmed bandwidth limit. Other embodiments are described and claimed.
The present application claims the benefit Indian Provisional Patent Application No. 201741030632, filed Aug. 30, 2017, and U.S. Provisional Patent Application No. 62/584,401, filed Nov. 10, 2017.
BACKGROUNDDatacenters and other large computer networks typically include multiple layers of switches. For example, servers may be installed in racks, and each server in a rack may be connected to a top-of-rack switch. Multiple top-of-rack switches may be connected to an upstream switch, and so on. Therefore, communicating between servers or other nodes in different racks may require traversing multiple switch layers. Traversing each layer of switches may introduce queuing latency.
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.
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.
Referring now to
Referring now to
It should be appreciated that each of the other pods 120, 130, 140 (as well as any additional pods of the data center 100) may be similarly structured as, and have components similar to, the pod 110 shown in and described in regard to
Referring now to
In the illustrative embodiments, each sled of the data center 100 is embodied as a chassis-less sled. That is, each sled has a chassis-less circuit board substrate on which physical resources (e.g., processors, memory, accelerators, storage, etc.) are mounted as discussed in more detail below. As such, the rack 240 is configured to receive the chassis-less sleds. For example, each pair 310 of elongated support arms 312 defines a sled slot 320 of the rack 240, which is configured to receive a corresponding chassis-less sled. To do so, each illustrative elongated support arm 312 includes a circuit board guide 330 configured to receive the chassis-less circuit board substrate of the sled. Each circuit board guide 330 is secured to, or otherwise mounted to, a top side 332 of the corresponding elongated support arm 312. For example, in the illustrative embodiment, each circuit board guide 330 is mounted at a distal end of the corresponding elongated support arm 312 relative to the corresponding elongated support post 302, 304. For clarity of the Figures, not every circuit board guide 330 may be referenced in each Figure.
Each circuit board guide 330 includes an inner wall that defines a circuit board slot 380 configured to receive the chassis-less circuit board substrate of a sled 400 when the sled 400 is received in the corresponding sled slot 320 of the rack 240. To do so, as shown in
It should be appreciated that each circuit board guide 330 is dual sided. That is, each circuit board guide 330 includes an inner wall that defines a circuit board slot 380 on each side of the circuit board guide 330. In this way, each circuit board guide 330 can support a chassis-less circuit board substrate on either side. As such, a single additional elongated support post may be added to the rack 240 to turn the rack 240 into a two-rack solution that can hold twice as many sled slots 320 as shown in
In some embodiments, various interconnects may be routed upwardly or downwardly through the elongated support posts 302, 304. To facilitate such routing, each elongated support post 302, 304 includes an inner wall that defines an inner chamber in which the interconnect may be located. The interconnects routed through the elongated support posts 302, 304 may be embodied as any type of interconnects including, but not limited to, data or communication interconnects to provide communication connections to each sled slot 320, power interconnects to provide power to each sled slot 320, and/or other types of interconnects.
The rack 240, in the illustrative embodiment, includes a support platform on which a corresponding optical data connector (not shown) is mounted. Each optical data connector is associated with a corresponding sled slot 320 and is configured to mate with an optical data connector of a corresponding sled 400 when the sled 400 is received in the corresponding sled slot 320. In some embodiments, optical connections between components (e.g., sleds, racks, and switches) in the data center 100 are made with a blind mate optical connection. For example, a door on each cable may prevent dust from contaminating the fiber inside the cable. In the process of connecting to a blind mate optical connector mechanism, the door is pushed open when the end of the cable enters the connector mechanism. Subsequently, the optical fiber inside the cable enters a gel within the connector mechanism and the optical fiber of one cable comes into contact with the optical fiber of another cable within the gel inside the connector mechanism.
The illustrative rack 240 also includes a fan array 370 coupled to the cross-support arms of the rack 240. The fan array 370 includes one or more rows of cooling fans 372, which are aligned in a horizontal line between the elongated support posts 302, 304. In the illustrative embodiment, the fan array 370 includes a row of cooling fans 372 for each sled slot 320 of the rack 240. As discussed above, each sled 400 does not include any on-board cooling system in the illustrative embodiment and, as such, the fan array 370 provides cooling for each sled 400 received in the rack 240. Each rack 240, in the illustrative embodiment, also includes a power supply associated with each sled slot 320. Each power supply is secured to one of the elongated support arms 312 of the pair 310 of elongated support arms 312 that define the corresponding sled slot 320. For example, the rack 240 may include a power supply coupled or secured to each elongated support arm 312 extending from the elongated support post 302. Each power supply includes a power connector configured to mate with a power connector of the sled 400 when the sled 400 is received in the corresponding sled slot 320. In the illustrative embodiment, the sled 400 does not include any on-board power supply and, as such, the power supplies provided in the rack 240 supply power to corresponding sleds 400 when mounted to the rack 240.
Referring now to
As discussed above, the illustrative sled 400 includes a chassis-less circuit board substrate 602, which supports various physical resources (e.g., electrical components) mounted thereon. It should be appreciated that the circuit board substrate 602 is “chassis-less” in that the sled 400 does not include a housing or enclosure. Rather, the chassis-less circuit board substrate 602 is open to the local environment. The chassis-less circuit board substrate 602 may be formed from any material capable of supporting the various electrical components mounted thereon. For example, in an illustrative embodiment, the chassis-less circuit board substrate 602 is formed from an FR-4 glass-reinforced epoxy laminate material. Of course, other materials may be used to form the chassis-less circuit board substrate 602 in other embodiments.
As discussed in more detail below, the chassis-less circuit board substrate 602 includes multiple features that improve the thermal cooling characteristics of the various electrical components mounted on the chassis-less circuit board substrate 602. As discussed, the chassis-less circuit board substrate 602 does not include a housing or enclosure, which may improve the airflow over the electrical components of the sled 400 by reducing those structures that may inhibit air flow. For example, because the chassis-less circuit board substrate 602 is not positioned in an individual housing or enclosure, there is no backplane (e.g., a backplate of the chassis) to the chassis-less circuit board substrate 602, which could inhibit air flow across the electrical components. Additionally, the chassis-less circuit board substrate 602 has a geometric shape configured to reduce the length of the airflow path across the electrical components mounted to the chassis-less circuit board substrate 602. For example, the illustrative chassis-less circuit board substrate 602 has a width 604 that is greater than a depth 606 of the chassis-less circuit board substrate 602. In one particular embodiment, for example, the chassis-less circuit board substrate 602 has a width of about 21 inches and a depth of about 9 inches, compared to a typical server that has a width of about 17 inches and a depth of about 39 inches. As such, an airflow path 608 that extends from a front edge 610 of the chassis-less circuit board substrate 602 toward a rear edge 612 has a shorter distance relative to typical servers, which may improve the thermal cooling characteristics of the sled 400. Furthermore, although not illustrated in
As discussed above, the illustrative sled 400 includes one or more physical resources 620 mounted to a top side 650 of the chassis-less circuit board substrate 602. Although two physical resources 620 are shown in
The sled 400 also includes one or more additional physical resources 630 mounted to the top side 650 of the chassis-less circuit board substrate 602. In the illustrative embodiment, the additional physical resources include a network interface controller (NIC) as discussed in more detail below. Of course, depending on the type and functionality of the sled 400, the physical resources 630 may include additional or other electrical components, circuits, and/or devices in other embodiments.
The physical resources 620 are communicatively coupled to the physical resources 630 via an input/output (I/O) subsystem 622. The I/O subsystem 622 may be embodied as circuitry and/or components to facilitate input/output operations with the physical resources 620, the physical resources 630, and/or other components of the sled 400. For example, the I/O subsystem 622 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 the illustrative embodiment, the I/O subsystem 622 is embodied as, or otherwise includes, a double data rate 4 (DDR4) data bus or a DDRS data bus.
In some embodiments, the sled 400 may also include a resource-to-resource interconnect 624. The resource-to-resource interconnect 624 may be embodied as any type of communication interconnect capable of facilitating resource-to-resource communications. In the illustrative embodiment, the resource-to-resource interconnect 624 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the resource-to-resource interconnect 624 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to resource-to-resource communications.
The sled 400 also includes a power connector 640 configured to mate with a corresponding power connector of the rack 240 when the sled 400 is mounted in the corresponding rack 240. The sled 400 receives power from a power supply of the rack 240 via the power connector 640 to supply power to the various electrical components of the sled 400. That is, the sled 400 does not include any local power supply (i.e., an on-board power supply) to provide power to the electrical components of the sled 400. The exclusion of a local or on-board power supply facilitates the reduction in the overall footprint of the chassis-less circuit board substrate 602, which may increase the thermal cooling characteristics of the various electrical components mounted on the chassis-less circuit board substrate 602 as discussed above. In some embodiments, power is provided to the processors 820 through vias directly under the processors 820 (e.g., through the bottom side 750 of the chassis-less circuit board substrate 602), providing an increased thermal budget, additional current and/or voltage, and better voltage control over typical boards.
In some embodiments, the sled 400 may also include mounting features 642 configured to mate with a mounting arm, or other structure, of a robot to facilitate the placement of the sled 600 in a rack 240 by the robot. The mounting features 642 may be embodied as any type of physical structures that allow the robot to grasp the sled 400 without damaging the chassis-less circuit board substrate 602 or the electrical components mounted thereto. For example, in some embodiments, the mounting features 642 may be embodied as non-conductive pads attached to the chassis-less circuit board substrate 602. In other embodiments, the mounting features may be embodied as brackets, braces, or other similar structures attached to the chassis-less circuit board substrate 602. The particular number, shape, size, and/or make-up of the mounting feature 642 may depend on the design of the robot configured to manage the sled 400.
Referring now to
The memory devices 720 may be embodied as any type of memory device capable of storing data for the physical resources 620 during operation of the sled 400, such as any type of volatile (e.g., dynamic random access memory (DRAM), etc.) or non-volatile memory. 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 next-generation nonvolatile devices, such as Intel 3D XPoint™ memory 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, the memory device 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.
Referring now to
In the illustrative compute sled 800, the physical resources 620 are embodied as processors 820. Although only two processors 820 are shown in
In some embodiments, the compute sled 800 may also include a processor-to-processor interconnect 842. Similar to the resource-to-resource interconnect 624 of the sled 400 discussed above, the processor-to-processor interconnect 842 may be embodied as any type of communication interconnect capable of facilitating processor-to-processor interconnect 842 communications. In the illustrative embodiment, the processor-to-processor interconnect 842 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the processor-to-processor interconnect 842 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to processor-to-processor communications.
The compute sled 800 also includes a communication circuit 830. The illustrative communication circuit 830 includes a network interface controller (NIC) 832, which may also be referred to as a host fabric interface (HFI). The NIC 832 may be embodied as, or otherwise include, any type of integrated circuit, discrete circuits, controller chips, chipsets, add-in-boards, daughtercards, network interface cards, other devices that may be used by the compute sled 800 to connect with another compute device (e.g., with other sleds 400). In some embodiments, the NIC 832 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 NIC 832 may include a local processor (not shown) and/or a local memory (not shown) that are both local to the NIC 832. In such embodiments, the local processor of the NIC 832 may be capable of performing one or more of the functions of the processors 820. Additionally or alternatively, in such embodiments, the local memory of the NIC 832 may be integrated into one or more components of the compute sled at the board level, socket level, chip level, and/or other levels.
The communication circuit 830 is communicatively coupled to an optical data connector 834. The optical data connector 834 is configured to mate with a corresponding optical data connector of the rack 240 when the compute sled 800 is mounted in the rack 240. Illustratively, the optical data connector 834 includes a plurality of optical fibers which lead from a mating surface of the optical data connector 834 to an optical transceiver 836. The optical transceiver 836 is configured to convert incoming optical signals from the rack-side optical data connector to electrical signals and to convert electrical signals to outgoing optical signals to the rack-side optical data connector. Although shown as forming part of the optical data connector 834 in the illustrative embodiment, the optical transceiver 836 may form a portion of the communication circuit 830 in other embodiments.
In some embodiments, the compute sled 800 may also include an expansion connector 840. In such embodiments, the expansion connector 840 is configured to mate with a corresponding connector of an expansion chassis-less circuit board substrate to provide additional physical resources to the compute sled 800. The additional physical resources may be used, for example, by the processors 820 during operation of the compute sled 800. The expansion chassis-less circuit board substrate may be substantially similar to the chassis-less circuit board substrate 602 discussed above and may include various electrical components mounted thereto. The particular electrical components mounted to the expansion chassis-less circuit board substrate may depend on the intended functionality of the expansion chassis-less circuit board substrate. For example, the expansion chassis-less circuit board substrate may provide additional compute resources, memory resources, and/or storage resources. As such, the additional physical resources of the expansion chassis-less circuit board substrate may include, but is not limited to, processors, memory devices, storage devices, and/or accelerator circuits including, for example, field programmable gate arrays (FPGA), application-specific integrated circuits (ASICs), security co-processors, graphics processing units (GPUs), machine learning circuits, or other specialized processors, controllers, devices, and/or circuits.
Referring now to
As discussed above, the individual processors 820 and communication circuit 830 are mounted to the top side 650 of the chassis-less circuit board substrate 602 such that no two heat-producing, electrical components shadow each other. In the illustrative embodiment, the processors 820 and communication circuit 830 are mounted in corresponding locations on the top side 650 of the chassis-less circuit board substrate 602 such that no two of those physical resources are linearly in-line with others along the direction of the airflow path 608. It should be appreciated that, although the optical data connector 834 is in-line with the communication circuit 830, the optical data connector 834 produces no or nominal heat during operation.
The memory devices 720 of the compute sled 800 are mounted to the bottom side 750 of the of the chassis-less circuit board substrate 602 as discussed above in regard to the sled 400. Although mounted to the bottom side 750, the memory devices 720 are communicatively coupled to the processors 820 located on the top side 650 via the I/O subsystem 622. Because the chassis-less circuit board substrate 602 is embodied as a double-sided circuit board, the memory devices 720 and the processors 820 may be communicatively coupled by one or more vias, connectors, or other mechanisms extending through the chassis-less circuit board substrate 602. Of course, each processor 820 may be communicatively coupled to a different set of one or more memory devices 720 in some embodiments. Alternatively, in other embodiments, each processor 820 may be communicatively coupled to each memory device 720. In some embodiments, the memory devices 720 may be mounted to one or more memory mezzanines on the bottom side of the chassis-less circuit board substrate 602 and may interconnect with a corresponding processor 820 through a ball-grid array.
Each of the processors 820 includes a heatsink 850 secured thereto. Due to the mounting of the memory devices 720 to the bottom side 750 of the chassis-less circuit board substrate 602 (as well as the vertical spacing of the sleds 400 in the corresponding rack 240), the top side 650 of the chassis-less circuit board substrate 602 includes additional “free” area or space that facilitates the use of heatsinks 850 having a larger size relative to traditional heatsinks used in typical servers. Additionally, due to the improved thermal cooling characteristics of the chassis-less circuit board substrate 602, none of the processor heatsinks 850 include cooling fans attached thereto. That is, each of the heatsinks 850 is embodied as a fan-less heatsinks.
Referring now to
In the illustrative accelerator sled 1000, the physical resources 620 are embodied as accelerator circuits 1020. Although only two accelerator circuits 1020 are shown in
In some embodiments, the accelerator sled 1000 may also include an accelerator-to-accelerator interconnect 1042. Similar to the resource-to-resource interconnect 624 of the sled 600 discussed above, the accelerator-to-accelerator interconnect 1042 may be embodied as any type of communication interconnect capable of facilitating accelerator-to-accelerator communications. In the illustrative embodiment, the accelerator-to-accelerator interconnect 1042 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the accelerator-to-accelerator interconnect 1042 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to processor-to-processor communications. In some embodiments, the accelerator circuits 1020 may be daisy-chained with a primary accelerator circuit 1020 connected to the NIC 832 and memory 720 through the I/O subsystem 622 and a secondary accelerator circuit 1020 connected to the NIC 832 and memory 720 through a primary accelerator circuit 1020.
Referring now to
Referring now to
In the illustrative storage sled 1200, the physical resources 620 are embodied as storage controllers 1220. Although only two storage controllers 1220 are shown in
In some embodiments, the storage sled 1200 may also include a controller-to-controller interconnect 1242. Similar to the resource-to-resource interconnect 624 of the sled 400 discussed above, the controller-to-controller interconnect 1242 may be embodied as any type of communication interconnect capable of facilitating controller-to-controller communications. In the illustrative embodiment, the controller-to-controller interconnect 1242 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the controller-to-controller interconnect 1242 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to processor-to-processor communications.
Referring now to
The storage cage 1252 illustratively includes sixteen mounting slots 1256 and is capable of mounting and storing sixteen solid state drives 1254. Of course, the storage cage 1252 may be configured to store additional or fewer solid state drives 1254 in other embodiments. Additionally, in the illustrative embodiment, the solid state drivers are mounted vertically in the storage cage 1252, but may be mounted in the storage cage 1252 in a different orientation in other embodiments. Each solid state drive 1254 may be embodied as any type of data storage device capable of storing long term data. To do so, the solid state drives 1254 may include volatile and non-volatile memory devices discussed above.
As shown in
As discussed above, the individual storage controllers 1220 and the communication circuit 830 are mounted to the top side 650 of the chassis-less circuit board substrate 602 such that no two heat-producing, electrical components shadow each other. For example, the storage controllers 1220 and the communication circuit 830 are mounted in corresponding locations on the top side 650 of the chassis-less circuit board substrate 602 such that no two of those electrical components are linearly in-line with other along the direction of the airflow path 608.
The memory devices 720 of the storage sled 1200 are mounted to the bottom side 750 of the of the chassis-less circuit board substrate 602 as discussed above in regard to the sled 400. Although mounted to the bottom side 750, the memory devices 720 are communicatively coupled to the storage controllers 1220 located on the top side 650 via the I/O subsystem 622. Again, because the chassis-less circuit board substrate 602 is embodied as a double-sided circuit board, the memory devices 720 and the storage controllers 1220 may be communicatively coupled by one or more vias, connectors, or other mechanisms extending through the chassis-less circuit board substrate 602. Each of the storage controllers 1220 includes a heatsink 1270 secured thereto. As discussed above, due to the improved thermal cooling characteristics of the chassis-less circuit board substrate 602 of the storage sled 1200, none of the heatsinks 1270 include cooling fans attached thereto. That is, each of the heatsinks 1270 is embodied as a fan-less heatsink.
Referring now to
In the illustrative memory sled 1400, the physical resources 620 are embodied as memory controllers 1420. Although only two memory controllers 1420 are shown in
In some embodiments, the memory sled 1400 may also include a controller-to-controller interconnect 1442. Similar to the resource-to-resource interconnect 624 of the sled 400 discussed above, the controller-to-controller interconnect 1442 may be embodied as any type of communication interconnect capable of facilitating controller-to-controller communications. In the illustrative embodiment, the controller-to-controller interconnect 1442 is embodied as a high-speed point-to-point interconnect (e.g., faster than the I/O subsystem 622). For example, the controller-to-controller interconnect 1442 may be embodied as a QuickPath Interconnect (QPI), an UltraPath Interconnect (UPI), or other high-speed point-to-point interconnect dedicated to processor-to-processor communications. As such, in some embodiments, a memory controller 1420 may access, through the controller-to-controller interconnect 1442, memory that is within the memory set 1432 associated with another memory controller 1420. In some embodiments, a scalable memory controller is made of multiple smaller memory controllers, referred to herein as “chiplets”, on a memory sled (e.g., the memory sled 1400). The chiplets may be interconnected (e.g., using EMIB (Embedded Multi-Die Interconnect Bridge)). The combined chiplet memory controller may scale up to a relatively large number of memory controllers and I/O ports, (e.g., up to 16 memory channels). In some embodiments, the memory controllers 1420 may implement a memory interleave (e.g., one memory address is mapped to the memory set 1430, the next memory address is mapped to the memory set 1432, and the third address is mapped to the memory set 1430, etc.). The interleaving may be managed within the memory controllers 1420, or from CPU sockets (e.g., of the compute sled 800) across network links to the memory sets 1430, 1432, and may improve the latency associated with performing memory access operations as compared to accessing contiguous memory addresses from the same memory device.
Further, in some embodiments, the memory sled 1400 may be connected to one or more other sleds 400 (e.g., in the same rack 240 or an adjacent rack 240) through a waveguide, using the waveguide connector 1480. In the illustrative embodiment, the waveguides are 64 millimeter waveguides that provide 16 Rx (i.e., receive) lanes and 16 Rt (i.e., transmit) lanes. Each lane, in the illustrative embodiment, is either 16 Ghz or 32 Ghz. In other embodiments, the frequencies may be different. Using a waveguide may provide high throughput access to the memory pool (e.g., the memory sets 1430, 1432) to another sled (e.g., a sled 400 in the same rack 240 or an adjacent rack 240 as the memory sled 1400) without adding to the load on the optical data connector 834.
Referring now to
Additionally, in some embodiments, the orchestrator server 1520 may identify trends in the resource utilization of the workload (e.g., the application 1532), such as by identifying phases of execution (e.g., time periods in which different operations, each having different resource utilizations characteristics, are performed) of the workload (e.g., the application 1532) and pre-emptively identifying available resources in the data center 100 and allocating them to the managed node 1570 (e.g., within a predefined time period of the associated phase beginning). In some embodiments, the orchestrator server 1520 may model performance based on various latencies and a distribution scheme to place workloads among compute sleds and other resources (e.g., accelerator sleds, memory sleds, storage sleds) in the data center 100. For example, the orchestrator server 1520 may utilize a model that accounts for the performance of resources on the sleds 400 (e.g., FPGA performance, memory access latency, etc.) and the performance (e.g., congestion, latency, bandwidth) of the path through the network to the resource (e.g., FPGA). As such, the orchestrator server 1520 may determine which resource(s) should be used with which workloads based on the total latency associated with each potential resource available in the data center 100 (e.g., the latency associated with the performance of the resource itself in addition to the latency associated with the path through the network between the compute sled executing the workload and the sled 400 on which the resource is located).
In some embodiments, the orchestrator server 1520 may generate a map of heat generation in the data center 100 using telemetry data (e.g., temperatures, fan speeds, etc.) reported from the sleds 400 and allocate resources to managed nodes as a function of the map of heat generation and predicted heat generation associated with different workloads, to maintain a target temperature and heat distribution in the data center 100. Additionally or alternatively, in some embodiments, the orchestrator server 1520 may organize received telemetry data into a hierarchical model that is indicative of a relationship between the managed nodes (e.g., a spatial relationship such as the physical locations of the resources of the managed nodes within the data center 100 and/or a functional relationship, such as groupings of the managed nodes by the customers the managed nodes provide services for, the types of functions typically performed by the managed nodes, managed nodes that typically share or exchange workloads among each other, etc.). Based on differences in the physical locations and resources in the managed nodes, a given workload may exhibit different resource utilizations (e.g., cause a different internal temperature, use a different percentage of processor or memory capacity) across the resources of different managed nodes. The orchestrator server 1520 may determine the differences based on the telemetry data stored in the hierarchical model and factor the differences into a prediction of future resource utilization of a workload if the workload is reassigned from one managed node to another managed node, to accurately balance resource utilization in the data center 100.
To reduce the computational load on the orchestrator server 1520 and the data transfer load on the network, in some embodiments, the orchestrator server 1520 may send self- test information to the sleds 400 to enable each sled 400 to locally (e.g., on the sled 400) determine whether telemetry data generated by the sled 400 satisfies one or more conditions (e.g., an available capacity that satisfies a predefined threshold, a temperature that satisfies a predefined threshold, etc.). Each sled 400 may then report back a simplified result (e.g., yes or no) to the orchestrator server 1520, which the orchestrator server 1520 may utilize in determining the allocation of resources to managed nodes.
Referring now to
The network elements of the system 1600 are organized into a network topology. As shown, the sleds 1610, 1612, 1614 may be organized into a rack and each connected to the switch 1606, which may be embodied as a top-of-rack switch, middle-of-rack switch, end-of-row switch, or other switch device. Similarly, the sleds 1616, 1618, 1620 are connected to the switch 1608. The switches 1606, 1608 are in turn connected to the switch 1604, which may be embodied as a data center domain switch or other upstream switch. The resource manager server 1602 is illustrated as being connected to the upstream switch 1604, however, in other embodiments it may be connected to any other location in the network topology. Thus, as shown in
Additionally, although the resource manager server 1602 is illustrated as a single server computing device, in some embodiments, the resource manager server 1602 may be embodied as a “virtual server” formed from multiple computing devices distributed across the system 1600 and/or operating in a public or private cloud. Accordingly, although the resource manager server 1602 is illustrated in
In use, as described further below, the resource manager server 1602 may discover the network topology of the system 1600 and construct a model of resource oversubscription in the system 1600. Resource oversubscription may include network uplink oversubscription, storage resource oversubscription, or any other circumstance in which the aggregate bandwidth or other demand generated by the sleds exceeds the available bandwidth or other capacity of the system 1600. The resource manager server 1602 may determine bandwidth limits for each sled or other network element of the system 1600 and program those bandwidth limits to the network elements. Each network element enforces the programmed bandwidth limits, which may reduce network congestion. By reducing congestion, the system 1600 may eliminate or reduce queuing latency for each layer of switches. For example, bandwidth limits for a storage sled 1610 may ensure that non-volatile memory express (NVMe) over Ethernet data traffic generated by the storage sled 1610 (and other storage sleds) does not exceed available upstream bandwidth. Accordingly, the system 1600 may improve network latency and reduce network congestion, without implementing expensive bandwidth reservation mechanisms at the switch level.
Referring now to
The processor 1720 may be embodied as any type of processor capable of performing the functions described herein. For example, the processor 1720 may be embodied as a single or multi-core processor(s), digital signal processor, microcontroller, or other processor or processing/controlling circuit. Similarly, the memory 1724 may be embodied as any type of volatile or non-volatile memory or data storage capable of performing the functions described herein. In operation, the memory 1724 may store various data and software used during operation of the computing device 1700 such operating systems, applications, programs, libraries, and drivers. The memory 1724 is communicatively coupled to the processor 1720 via the I/O subsystem 1722, which may be embodied as circuitry and/or components to facilitate input/output operations with the processor 1720, the memory 1724, and other components of the computing device 1700. For example, the I/O subsystem 1722 may be embodied as, or otherwise include, memory controller hubs, input/output control hubs, sensor hubs, firmware devices, communication links (i.e., 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 1722 may form a portion of a system-on-a-chip (SoC) and be incorporated, along with the processor 1720, the memory 1724, and other components of the computing device 1700, on a single integrated circuit chip.
The data storage device 1726 may be embodied as any type of device or 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, non-volatile flash memory, or other data storage devices. The computing device 1700 may also include a communications subsystem 1728, which may be embodied as any communication circuit, device, or collection thereof, capable of enabling communications between the computing device 1700 and other remote devices over a computer network (not shown). The communications subsystem 1728 may be configured to use any one or more communication technology (e.g., wired or wireless communications) and associated protocols (e.g., Ethernet, InfiniBand®, Bluetooth®, Wi-Fi®, WiMAX, 3G, 4G LTE, etc.) to effect such communication. As shown, the communication subsystem 1728 may include a network interface controller (NIC) 1330.
The illustrative communications subsystem 1728 includes a network interface controller (NIC) 1330. The NIC 1730 may be embodied as one or more add-in-boards, daughtercards, controller chips, chipsets, circuits, or other devices that may be used by the computing device 1700 for network communications with remote devices. For example, the NIC 1730 may be embodied as an expansion card coupled to the I/O subsystem 1722 over an expansion bus such as PCI Express. As another example, in some embodiments the NIC 1730 may be embodied as a network controller, host fabric interface, or other component integrated with the I/O subsystem 1722, the processor 1720, an SoC, and/or one or more other components of the computing device 1700.
As shown, the computing device 1700 may also include one or more peripheral devices 1732. The peripheral devices 1732 may include any number of additional input/output devices, interface devices, and/or other peripheral devices. For example, in some embodiments, the peripheral devices 1732 may include a display, touch screen, graphics circuitry, keyboard, mouse, speaker system, microphone, network interface, and/or other input/output devices, interface devices, and/or peripheral devices.
Referring now to
The topology manager 1802 is configured to discover the topology of the sleds coupled to a layer of switches that are communicatively coupled to the resource manager server 1602. The model constructor 1804 is configured to construct a model of network connectivity between the plurality of sleds and the layer of switches based on the topology. Constructing the model may include identifying which sleds of the plurality of sleds are connected to a particular switch of the layer of switches.
The bandwidth limit determiner 1806 is configured to determine an oversubscription of the network based on the model of network connectivity. The oversubscription is based on an available bandwidth for the layer of switches and a maximum bandwidth of the sleds. The bandwidth limit determiner 1806 may determine a network uplink oversubscription for the layer of switches or may determine a storage resource oversubscription of the sleds. The bandwidth limit determiner 1806 is further configured to determine a bandwidth limit for each sled based on the oversubscription. The bandwidth limit programmer 1808 is configured to program each sled with the corresponding bandwidth limit. The bandwidth limit programmer 1808 may communicate the bandwidth limit to the NIC 1730 of the corresponding sled.
The utilization manager 1810 is configured to monitor a bandwidth utilization of the sleds. Monitoring the bandwidth utilization may include receiving telemetry data from the sleds that is indicative of the bandwidth utilized by each sled. The utilization manager 1810 is further configured to determine whether the network is congested based on the bandwidth utilization of the plurality of sleds. For example, determining whether the network is congested may include determining whether a queue depth of the network exceeds a predetermined queue depth limit for a predetermined amount of time. In some embodiments, determining whether the network is congested may include monitoring bandwidth per class of network traffic, such as NVMe over Ethernet traffic, field-programmable gate array (FPGA) over Ethernet traffic, storage traffic, and/or other traffic classes. Each traffic class may be independently monitored with its own queue depth controls. Further, bandwidth may be limited at the source or target, or in some embodiments based on source and target pair combinations. The utilization manager 1810 is further configured to modify a bandwidth limit in response to determining that the network is congested. The bandwidth limit may be reduced for each sled that is associated with a high input rate flow.
Referring now to
The bandwidth programmer 1906 is configured to receive a bandwidth limit for the sled from the resource manager server 1602 and to program the bandwidth limit to the NIC 1730 of the sled. The bandwidth limit manager 1904 is configured to enforce, by the NIC 1730, the bandwidth limit in response to programming the bandwidth limit.
The telemetry data manager 1908 is configured to send telemetry data indicative of utilization of the NIC 1730 to the resource manager server 1602. The telemetry data may be sent by the NIC 1730. The telemetry data may indicative of a NIC queue depth and/or or a network stack queue depth.
Referring now to
In block 2006, the resource manager server 1602 constructs a model of network connectivity between the components of the system 1600. The model may identify network connections and the associated available bandwidth between sleds, switches, and other network elements of the system 1600.
In block 2008, the resource manager server 1602 determines oversubscription of the system 1600 based on the model of network connectivity. As described above, the system 1600 may be organized in layers, and each layer may have a maximum amount of available bandwidth or other resources. Oversubscription may exist if the total maximum bandwidth or other resource demand of a layer exceeds the available bandwidth or other resources of a higher layer. In some embodiments, in block 2010, the resource manager server 1602 may determine network uplink oversubscription. For example, as shown in
In block 2014, the resource manager server 1602 determines bandwidth limits for each sled in the system 1600 based on the oversubscription. The bandwidth limits may be determined in order to prevent or reduce network congestion in the system 1600. For example, again referring to
In block 2016, the resource manager server 1602 programs each sled with the corresponding bandwidth limit. After being programmed, each sled enforces the bandwidth limits, as described further below in connection with
In block 2020, the resource manager server 1602 may receive bandwidth telemetry from the sleds of the system 1600. The bandwidth telemetry may indicate the current bandwidth usage of the sled and/or whether the associated network connection is congested. For example, the bandwidth telemetry may indicate queue depth of the NIC 1730, the associated switch port, and/or the networking stack of the sled.
In block 2022, the resource manager server 1602 identifies network congestion based on the telemetry. The resource manager server 1602 may use any appropriate algorithm to identify dropped packets, increased latency, or otherwise identify the network congestion. In some embodiments, in block 2024, the resource manager server 1602 may determine whether any queue depth in the system exceeds a predetermined threshold queue depth for longer than a predetermined time. For example, the resource manager server 1602 may analyze the queue depth of a NIC 1730, a switch port, and/or a networking stack of a sled. In block 2026, the resource manager server 1602 determines whether congestion has been detected. If not, the method 2000 loops back to block 2020 to continue monitoring network utilization. If congestion is detected, the method 2000 advances to block 2028.
In block 2028, the resource manager server 1602 modifies one or more bandwidth limits to reduce or eliminate the congestion. In some embodiments, in block 2030, the resource manager server 1602 may identify one or more high-input rate flows in the system 1600. For example, one or more storage sleds 1610 generating NVMe over Ethernet data may generate high-input rate flows. The resource manager server 1602 may reduce the input rate bandwidth limit associated with the high-input rate flows. Additionally or alternatively, in some embodiments the resource manager server 1602 may generate one or more alerts concerning the congestion, and a network administrator may provide modified bandwidth limits. In some embodiments, alternate network routes may be possible. If alternate routes are possible, based on the congestion telemetry data, different bandwidth limits may be set for different routes to reduce the congestion rate. After modifying the bandwidth limits, the method 2000 loops back to block 2016 to program the sleds with the modified bandwidth limits and continue monitoring network utilization.
Referring now to
In block 2104, the storage sled 1610 programs one or more network interface controllers (NICs) 1330 of the storage sled with the new bandwidth limit. After being programmed, the NIC 1730 may throttle or otherwise limit bandwidth used by the storage sled 1610 to below the bandwidth limit. In particular, in some embodiments in block 2106 the storage sled 1610 may set a maximum input bandwidth for the NIC 1730. Thus, the bandwidth limits may limit the amount of data (e.g., NVMe over Ethernet data) generated by the storage sled 1610 and submitted to the switch 1606. Although illustrated as being enforced by the NIC 1730, it should be understood that in some embodiments, the bandwidth limits may be enforced by other components of the storage sled 1610, such as an operating system, software networking stack, NVMe over Ethernet subsystem, or other component.
In block 2108, the storage sled 1610 determines whether to send telemetry data to the resource manager server 1602. For example, the storage sled 1610 may be configured by an administrator to send telemetry data. In some embodiments, the storage sled 1610 may send telemetry data in response to certain events, for example in response to detected network congestion. If the storage sled 1610 determines not to send telemetry data, the method 2100 loops back to block 2102 to continue monitoring for updated bandwidth limits. If the storage sled 1610 determines to send telemetry data, the method 2100 advances to block 2110.
In block 2110, the storage sled 1610 sends bandwidth telemetry data to the resource manager server 1602. As described above, the bandwidth telemetry may indicate the current bandwidth usage of the sled and/or whether the associated network connection is congested. For example, the bandwidth telemetry may indicate queue depth of the NIC 1730, the associated switch port, and/or the network stack of the sled. In some embodiments, in block 2112 the storage sled 1610 may retrieve the telemetry data from the NIC 1730 of the storage sled 1610. For example, an operating system, software networking stack, or other component of the storage sled 1610 may retrieve telemetry data from the NIC 1730. In some embodiments, in block 2114 the storage sled 1610 may send the telemetry data from the NIC 1730 to the resource manager server 1602. The NIC 1730 may send the telemetry data out-of-band or otherwise without the involvement of the operating system, software networking stack, or other components of the storage sled 1610. After sending the telemetry data, the method 2100 loops back to block 2102 to continue monitoring for updated bandwidth limits.
EXAMPLESIllustrative 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 resource manager server for bandwidth allocation, the resource manager server comprising: one or more processors; and one or more memory devices having stored therein a plurality of instructions that, when executed by the one or more processors, cause the resource manager server to: discover a topology of a plurality of sleds coupled to a layer of switches that are communicatively coupled to the resource manager server; construct a model of network connectivity between the plurality of sleds and the layer of switches based on the topology; determine an oversubscription of a network based on the model of network connectivity, wherein the oversubscription is based on an available bandwidth for the layer of switches and a maximum bandwidth of the plurality of sleds; determine a bandwidth limit for each sled of the plurality of sleds based on the oversubscription; and program each sled of the plurality of sleds with the corresponding bandwidth limit.
Example 2 includes the subject matter of Example 1, and wherein to construct the model of network connectivity comprises to identify which sleds of the plurality of sleds are connected to a particular switch of the layer of switches.
Example 3 includes the subject matter of any of Examples 1 and 2, and wherein to determine the oversubscription comprises to determine a network uplink oversubscription for the layer of switches.
Example 4 includes the subject matter of any of Examples 1-3, and wherein to determine the oversubscription comprises to determine a storage resource oversubscription of the plurality of sleds.
Example 5 includes the subject matter of any of Examples 1-4, and wherein to program the bandwidth limit for each sled comprises to communicate the bandwidth limit to a network interface controller of the corresponding sled.
Example 6 includes the subject matter of any of Examples 1-5, and wherein the one or more memory devices have stored therein a plurality of instructions that, when executed by the one or more processors, further cause the resource manager server to: monitor a bandwidth utilization of the plurality of sleds; determine whether the network is congested based on the bandwidth utilization of the plurality of sleds; and modify a bandwidth limit in response to a determination that the network is congested.
Example 7 includes the subject matter of any of Examples 1-6, and wherein to monitor the bandwidth utilization of the plurality of sleds comprises to receive telemetry data from the plurality of sleds indicative of the bandwidth utilized by each sled.
Example 8 includes the subject matter of any of Examples 1-7, and wherein to determine whether the network is congested comprises to determine whether a queue depth of the network exceeds a predetermined queue depth limit for a predetermined amount of time.
Example 9 includes the subject matter of any of Examples 1-8, and wherein the queue depth comprises a switch port queue depth, a network interface controller queue depth, or a network stack queue depth.
Example 10 includes the subject matter of any of Examples 1-9, and wherein to modify the bandwidth limit comprises to: identify a first sled of the plurality of sleds associated with a high input rate flow; and reduce an input rate of the bandwidth limit for the first sled.
Example 11 includes a sled for bandwidth allocation, the sled communicatively coupled to a layer of switches that are communicatively coupled to a resource manager server on a network, the sled comprising: one or more processors; and one or more memory devices having stored therein a plurality of instructions that, when executed by the one or more processors, cause the sled to: receive a bandwidth limit for the sled from the resource manager server; program the bandwidth limit to a network interface controller of the sled; and enforce, by the network interface controller, the bandwidth limit in response to programming of the bandwidth limit.
Example 12 includes the subject matter of Example 11, and wherein the one or more memory devices have stored therein a plurality of instructions that, when executed by the one or more processors, further cause the sled to send telemetry data indicative of a utilization of the network interface controller to the resource manager server of the network.
Example 13 includes the subject matter of any of Examples 11 and 12, and wherein to send the telemetry data comprises to send the telemetry data by the network interface controller.
Example 14 includes the subject matter of any of Examples 11-13, and wherein the telemetry data is indicative of a network interface controller queue depth, or a network stack queue depth.
Example 15 includes a method for bandwidth allocation, the method comprising: discovering, by a resource manager server of a network, a topology of a plurality of sleds coupled to a layer of switches that are communicatively coupled to the resource manager server; constructing, by the resource manager server, a model of network connectivity between the plurality of sleds and the layer of switches based on the topology; determining, by the resource manager server, an oversubscription of the network based on the model of network connectivity, wherein the oversubscription is based on an available bandwidth for the layer of switches and a maximum bandwidth of the plurality of sleds; determining, by the resource manager server, a bandwidth limit for each sled of the plurality of sleds based on the oversubscription; and programming, by the resource manager server, each sled of the plurality of sleds with the corresponding bandwidth limit.
Example 16 includes the subject matter of Example 15, and wherein constructing the model of network connectivity comprises identifying which sleds of the plurality of sleds are connected to a particular switch of the layer of switches.
Example 17 includes the subject matter of any of Examples 15 and 16, and wherein determining the oversubscription comprises determining a network uplink oversubscription for the layer of switches.
Example 18 includes the subject matter of any of Examples 15-17, and wherein determining the oversubscription comprises determining a storage resource oversubscription of the plurality of sleds.
Example 19 includes the subject matter of any of Examples 15-18, and wherein programming the bandwidth limit for each sled comprises communicating the bandwidth limit to a network interface controller of the corresponding sled.
Example 20 includes the subject matter of any of Examples 15-19, and further comprising: monitoring, by the resource manager server, a bandwidth utilization of the plurality of sleds; determining, by the resource manager server, whether the network is congested based on the bandwidth utilization of the plurality of sleds; and modifying, by the resource manager server, a bandwidth limit in response to determining that the network is congested.
Example 21 includes the subject matter of any of Examples 15-20, and wherein monitoring the bandwidth utilization of the plurality of sleds comprises receiving telemetry data from the plurality of sleds indicative of the bandwidth utilized by each sled.
Example 22 includes the subject matter of any of Examples 15-21, and wherein determining whether the network is congested comprises determining whether a queue depth of the network exceeds a predetermined queue depth limit for a predetermined amount of time.
Example 23 includes the subject matter of any of Examples 15-22, and wherein the queue depth comprises a switch port queue depth, a network interface controller queue depth, or a network stack queue depth.
Example 24 includes the subject matter of any of Examples 15-23, and wherein modifying the bandwidth limit comprises: identifying a first sled of the plurality of sleds associated with a high input rate flow; and reducing an input rate of the bandwidth limit for the first sled.
Example 25 includes a method for bandwidth allocation, the method comprising: receiving, by a sled of a plurality of sleds communicatively coupled to a layer of switches that are communicatively coupled to a resource manager server in a network, a bandwidth limit for the sled from the resource manager server; programming, by the sled, the bandwidth limit to a network interface controller of the sled; and enforcing, by the network interface controller of the sled, the bandwidth limit in response to programming the bandwidth limit.
Example 26 includes the subject matter of Example 25, and further comprising sending, by the sled, telemetry data indicative of a utilization of the network interface controller to the resource manager server of the network.
Example 27 includes the subject matter of any of Examples 25 and 26, and wherein sending the telemetry data comprises sending the telemetry data by the network interface controller.
Example 28 includes the subject matter of any of Examples 25-27, and wherein the telemetry data is indicative of a network interface controller queue depth, or a network stack queue depth.
Example 29 includes a computing device comprising: a processor; and a memory having stored therein a plurality of instructions that when executed by the processor cause the computing device to perform the method of any of Examples 15-28.
Example 30 includes one or more non-transitory, computer readable storage media comprising a plurality of instructions stored thereon that in response to being executed result in a computing device performing the method of any of Examples 15-28.
Example 31 includes a computing device comprising means for performing the method of any of Examples 15-28.
Example 32 includes a resource manager server for bandwidth allocation, the resource manager server comprising: topology manager circuitry to discover a topology of a plurality of sleds coupled to a layer of switches that are communicatively coupled to the resource manager server; model constructer circuitry to construct a model of network connectivity between the plurality of sleds and the layer of switches based on the topology; bandwidth limit determiner circuitry to (i) determine an oversubscription of a network based on the model of network connectivity, wherein the oversubscription is based on an available bandwidth for the layer of switches and a maximum bandwidth of the plurality of sleds, and (ii) determine a bandwidth limit for each sled of the plurality of sleds based on the oversubscription; and bandwidth limit programmer circuitry to program each sled of the plurality of sleds with the corresponding bandwidth limit.
Example 33 includes the subject matter of Example 32, and wherein to construct the model of network connectivity comprises to identify which sleds of the plurality of sleds are connected to a particular switch of the layer of switches.
Example 34 includes the subject matter of any of Examples 32 and 33, and wherein to determine the oversubscription comprises to determine a network uplink oversubscription for the layer of switches.
Example 35 includes the subject matter of any of Examples 32-34, and wherein to determine the oversubscription comprises to determine a storage resource oversubscription of the plurality of sleds.
Example 36 includes the subject matter of any of Examples 32-35, and wherein to program the bandwidth limit for each sled comprises to communicate the bandwidth limit to a network interface controller of the corresponding sled.
Example 37 includes the subject matter of any of Examples 32-36, and further comprising utilization manager circuitry to: monitor a bandwidth utilization of the plurality of sleds; determine whether the network is congested based on the bandwidth utilization of the plurality of sleds; and modify a bandwidth limit in response to a determination that the network is congested.
Example 38 includes the subject matter of any of Examples 32-37, and wherein to monitor the bandwidth utilization of the plurality of sleds comprises to receive telemetry data from the plurality of sleds indicative of the bandwidth utilized by each sled.
Example 39 includes the subject matter of any of Examples 32-38, and wherein to determine whether the network is congested comprises to determine whether a queue depth of the network exceeds a predetermined queue depth limit for a predetermined amount of time.
Example 40 includes the subject matter of any of Examples 32-39, and wherein the queue depth comprises a switch port queue depth, a network interface controller queue depth, or a network stack queue depth.
Example 41 includes the subject matter of any of Examples 32-40, and wherein to modify the bandwidth limit comprises to: identify a first sled of the plurality of sleds associated with a high input rate flow; and reduce an input rate of the bandwidth limit for the first sled.
Example 42 includes a sled for bandwidth allocation, the sled communicatively coupled to a layer of switches that communicatively coupled to a resource manager server on a network, the sled comprising: bandwidth programmer circuitry to: (i) receive a bandwidth limit for the sled from the resource manager server, and (ii) program the bandwidth limit to a network interface controller of the sled; and bandwidth limit manager circuitry to enforce, by the network interface controller, the bandwidth limit in response to programming of the bandwidth limit.
Example 43 includes the subject matter of Example 42, and further comprising telemetry data manager circuitry to send telemetry data indicative of a utilization of the network interface controller to the resource manager server of the network.
Example 44 includes the subject matter of any of Examples 42 and 43, and wherein to send the telemetry data comprises to send the telemetry data by the network interface controller.
Example 45 includes the subject matter of any of Examples 42-44, and wherein the telemetry data is indicative of a network interface controller queue depth, or a network stack queue depth.
Example 46 includes a resource manager server for bandwidth allocation, the resource manager server comprising: means for discovering a topology of a plurality of sleds coupled to a layer of switches that are communicatively coupled to the resource manager server in a network; means for constructing a model of network connectivity between the plurality of sleds and the layer of switches based on the topology; means for determining an oversubscription of the network based on the model of network connectivity, wherein the oversubscription is based on an available bandwidth for the layer of switches and a maximum bandwidth of the plurality of sleds; means for determining a bandwidth limit for each sled of the plurality of sleds based on the oversubscription; and means for programming each sled of the plurality of sleds with the corresponding bandwidth limit.
Example 47 includes the subject matter of Example 46, and wherein the means for constructing the model of network connectivity comprises means for identifying which sleds of the plurality of sleds are connected to a particular switch of the layer of switches.
Example 48 includes the subject matter of any of Examples 46 and 47, and wherein the means for determining the oversubscription comprises means for determining a network uplink oversubscription for the layer of switches.
Example 49 includes the subject matter of any of Examples 46-48, and wherein the means for determining the oversubscription comprises means for determining a storage resource oversubscription of the plurality of sleds.
Example 50 includes the subject matter of any of Examples 46-49, and wherein the means for programming the bandwidth limit for each sled comprises circuitry for communicating the bandwidth limit to a network interface controller of the corresponding sled.
Example 51 includes the subject matter of any of Examples 46-50, and further comprising: means for monitoring a bandwidth utilization of the plurality of sleds; means for determining whether the network is congested based on the bandwidth utilization of the plurality of sleds; and means for modifying a bandwidth limit in response to determining that the network is congested.
Example 52 includes the subject matter of any of Examples 46-51, and wherein the means for monitoring the bandwidth utilization of the plurality of sleds comprises circuitry for receiving telemetry data from the plurality of sleds indicative of the bandwidth utilized by each sled.
Example 53 includes the subject matter of any of Examples 46-52, and wherein the means for determining whether the network is congested comprises means for determining whether a queue depth of the network exceeds a predetermined queue depth limit for a predetermined amount of time.
Example 54 includes the subject matter of any of Examples 46-53, and wherein the queue depth comprises a switch port queue depth, a network interface controller queue depth, or a network stack queue depth.
Example 55 includes the subject matter of any of Examples 46-54, and wherein the means for modifying the bandwidth limit comprises: means for identifying a first sled of the plurality of sleds associated with a high input rate flow; and means for reducing an input rate of the bandwidth limit for the first sled.
Example 56 includes a sled for bandwidth allocation, the sled communicatively coupled to a layer of switches that are communicatively coupled to a resource manager server on a network, the sled comprising: circuitry for receiving a bandwidth limit for the sled from the resource manager server; means for programming the bandwidth limit to a network interface controller of the sled; and means for enforcing, by the network interface controller of the sled, the bandwidth limit in response to programming the bandwidth limit.
Example 57 includes the subject matter of Example 56, and further comprising means for sending telemetry data indicative of a utilization of the network interface controller to the resource manager server of the network.
Example 58 includes the subject matter of any of Examples 56 and 57, and wherein the means for sending the telemetry data comprises means for sending the telemetry data by the network interface controller.
Example 59 includes the subject matter of any of Examples 56-58, and wherein the telemetry data is indicative of a network interface controller queue depth, or a network stack queue depth.
Claims
1. A resource manager server for bandwidth allocation, the resource manager server comprising:
- one or more processors; and
- one or more memory devices having stored therein a plurality of instructions that, when executed by the one or more processors, cause the resource manager server to: discover a topology of a plurality of sleds coupled to a layer of switches that are communicatively coupled to the resource manager server; construct a model of network connectivity between the plurality of sleds and the layer of switches based on the topology; determine an oversubscription of a network based on the model of network connectivity, wherein the oversubscription is based on an available bandwidth for the layer of switches and a maximum bandwidth of the plurality of sleds; determine a bandwidth limit for each sled of the plurality of sleds based on the oversubscription; and program each sled of the plurality of sleds with the corresponding bandwidth limit.
2. The resource manager server of claim 1, wherein to construct the model of network connectivity comprises to identify which sleds of the plurality of sleds are connected to a particular switch of the layer of switches.
3. The resource manager server of claim 1, wherein to determine the oversubscription comprises to determine a network uplink oversubscription for the layer of switches.
4. The resource manager server of claim 1, wherein to determine the oversubscription comprises to determine a storage resource oversubscription of the plurality of sleds.
5. The resource manager server of claim 1, wherein to program the bandwidth limit for each sled comprises to communicate the bandwidth limit to a network interface controller of the corresponding sled.
6. The resource manager server of claim 1, wherein the one or more memory devices have stored therein a plurality of instructions that, when executed by the one or more processors, further cause the resource manager server to:
- monitor a bandwidth utilization of the plurality of sleds;
- determine whether the network is congested based on the bandwidth utilization of the plurality of sleds; and
- modify a bandwidth limit in response to a determination that the network is congested.
7. The resource manager server of claim 6, wherein to monitor the bandwidth utilization of the plurality of sleds comprises to receive telemetry data from the plurality of sleds indicative of the bandwidth utilized by each sled.
8. The resource manager server of claim 7, wherein to determine whether the network is congested comprises to determine whether a queue depth of the network exceeds a predetermined queue depth limit for a predetermined amount of time.
9. The resource manager server of claim 8, wherein the queue depth comprises a switch port queue depth, a network interface controller queue depth, or a network stack queue depth.
10. The resource manager server of claim 6, wherein to modify the bandwidth limit comprises to:
- identify a first sled of the plurality of sleds associated with a high input rate flow; and
- reduce an input rate of the bandwidth limit for the first sled.
11. One or more computer-readable storage media comprising a plurality of instructions stored thereon that, in response to being executed, cause a resource manager server to:
- discover a topology of a plurality of sleds coupled to a layer of switches that are communicatively coupled to the resource manager server in a network;
- construct a model of network connectivity between the plurality of sleds and the layer of switches based on the topology;
- determine an oversubscription of the network based on the model of network connectivity, wherein the oversubscription is based on an available bandwidth for the layer of switches and a maximum bandwidth of the plurality of sleds;
- determine a bandwidth limit for each sled of the plurality of sleds based on the oversubscription; and
- program each sled of the plurality of sleds with the corresponding bandwidth limit.
12. The one or more computer-readable storage media of claim 11, wherein to construct the model of network connectivity comprises to identify which sleds of the plurality of sleds are connected to a particular switch of the layer of switches.
13. The one or more computer-readable storage media of claim 11, wherein to determine the oversubscription comprises to determine a network uplink oversubscription for the layer of switches.
14. The one or more computer-readable storage media of claim 11, wherein to determine the oversubscription comprises to determine a storage resource oversubscription of the plurality of sleds.
15. The one or more computer-readable storage media of claim 11, wherein to program the bandwidth limit for each sled comprises to communicate the bandwidth limit to a network interface controller of the corresponding sled.
16. The one or more computer-readable storage media of claim 11, further comprising a plurality of instructions stored thereon that, in response to being executed, cause the resource manager server to:
- monitor a bandwidth utilization of the plurality of sleds;
- determine whether the network is congested based on the bandwidth utilization of the plurality of sleds; and
- modify a bandwidth limit in response to determining that the network is congested.
17. The one or more computer-readable storage media of claim 16, wherein to monitor the bandwidth utilization of the plurality of sleds comprises to receive telemetry data from the plurality of sleds indicative of the bandwidth utilized by each sled.
18. The one or more computer-readable storage media of claim 17, wherein to determine whether the network is congested comprises to determine whether a queue depth of the network exceeds a predetermined queue depth limit for a predetermined amount of time.
19. The one or more computer-readable storage media of claim 18, wherein the queue depth comprises a switch port queue depth, a network interface controller queue depth, or a network stack queue depth.
20. The one or more computer-readable storage media of claim 16, wherein to modify the bandwidth limit comprises to:
- identify a first sled of the plurality of sleds associated with a high input rate flow; and
- reduce an input rate of the bandwidth limit for the first sled.
21. A resource manager server for bandwidth allocation, the resource manager server comprising:
- means for discovering a topology of a plurality of sleds coupled to a layer of switches that are communicatively coupled to the resource manager server in a network;
- means for constructing a model of network connectivity between the plurality of sleds and the layer of switches based on the topology;
- means for determining an oversubscription of the network based on the model of network connectivity, wherein the oversubscription is based on an available bandwidth for the layer of switches and a maximum bandwidth of the plurality of sleds;
- means for determining a bandwidth limit for each sled of the plurality of sleds based on the oversubscription; and
- means for programming each sled of the plurality of sleds with the corresponding bandwidth limit.
22. A method for bandwidth allocation, the method comprising:
- discovering, by a resource manager server of a network, a topology of a plurality of sleds coupled to a layer of switches that are communicatively coupled to the resource manager server;
- constructing, by the resource manager server, a model of network connectivity between the plurality of sleds and the layer of switches based on the topology;
- determining, by the resource manager server, an oversubscription of the network based on the model of network connectivity, wherein the oversubscription is based on an available bandwidth for the layer of switches and a maximum bandwidth of the plurality of sleds;
- determining, by the resource manager server, a bandwidth limit for each sled of the plurality of sleds based on the oversubscription; and
- programming, by the resource manager server, each sled of the plurality of sleds with the corresponding bandwidth limit.
23. The method of claim 22, further comprising:
- monitoring, by the resource manager server, a bandwidth utilization of the plurality of sleds;
- determining, by the resource manager server, whether the network is congested based on the bandwidth utilization of the plurality of sleds; and
- modifying, by the resource manage server, a bandwidth limit in response to determining that the network is congested.
24. The method of claim 23, wherein monitoring the bandwidth utilization of the plurality of sleds comprises receiving telemetry data from the plurality of sleds indicative of the bandwidth utilized by each sled.
25. The method of claim 24, wherein determining whether the network is congested comprises determining whether a queue depth of the network exceeds a predetermined queue depth limit for a predetermined amount of time.
Type: Application
Filed: Dec 29, 2017
Publication Date: Feb 28, 2019
Inventors: Mohan J. Kumar (Aloha, OR), Murugasamy K. Nachimuthu (Beaverton, OR)
Application Number: 15/858,288