TECHNOLOGIES FOR AUTO-DISCOVERY OF FAULT DOMAINS

BACKGROUND

A cloud computing facility, data center, server farm, or any similar large-scale computer architecture often includes a large number of computer devices. These computer devices may be organized in various ways. For example, based on connectivity or usage, these computer devices will often be organized into groups by network type, application type, or the like. Additionally, computer devices have specific power requirements and are connected to one or more power source devices. These power source devices may draw power from various sources and may not all be interconnected. As a result, groups of power source devices (sometimes referred to as power zones) may be separate from other groups of power source devices. In the event of a power outage or other power loss incident, the impact on one power source device may affect multiple connected computer devices. Known methods of maintaining an inventory of connections between computer devices, network devices, and power source devices include manual inventory of the computer devices that are connected to each power source device. This manual inventory may also be entered into a distributed data processing framework that performs data replication and other data management services for the computer devices.

BRIEF DESCRIPTION OF THE DRAWINGS

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

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

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

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

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

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

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

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

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

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

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

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

FIG. 12 is a simplified block diagram of at least one embodiment of a compute device for automatic discovery of fault domains;

FIG. 13 is a simplified block diagram of at least one embodiment of an environment established by the compute device of FIG. 12;

FIG. 14 is a simplified flow diagram of at least one embodiment of a method for automatic discovery of fault domains that may be executed by a sled shown in FIG. 16;

FIGS. 15 and 16 are a simplified flow diagram of at least one embodiment of a method for automatic discovery of fault domains that may be executed by the compute device of FIGS. 12-13; and

FIG. 17 is a diagram of a datacenter or computing network that includes one or more sleds and other compute devices, any one of which may perform a method for automatic discovery of fault domains as shown in FIGS. 15 and 16.

DETAILED DESCRIPTION OF THE DRAWINGS

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

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

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

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

FIG. 1 illustrates a conceptual overview of a data center 100 that may generally be representative of a data center or other type of computing network in/for which one or more techniques described herein may be implemented according to various embodiments. As shown in FIG. 1, data center 100 may generally contain a plurality of racks, each of which may house computing equipment comprising a respective set of physical resources. In the particular non-limiting example depicted in FIG. 1, data center 100 contains four racks 102A to 102D, which house computing equipment comprising respective sets of physical resources 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 dual inline memory modules (DIMMs), are located on a bottom side of the sled. As a result of the enhanced airflow provided by this design, the components may operate at higher frequencies and power levels than in typical systems, thereby increasing performance Furthermore, the sleds are configured to blindly mate with power and data communication cables in each rack 102A, 102B, 102C, 102D, enhancing their ability to be quickly removed, upgraded, reinstalled, and/or replaced. Similarly, individual components located on the sleds, such as processors, accelerators, memory, and data storage drives, are configured to be easily upgraded due to their increased spacing from each other. In the illustrative embodiment, the components additionally include hardware attestation features to prove their authenticity.

Furthermore, in the illustrative embodiment, the data center 100 utilizes a single network architecture (“fabric”) that supports multiple other network architectures including Ethernet and Omni-Path. The sleds, in the illustrative embodiment, are coupled to switches via optical fibers, which provide higher bandwidth and lower latency than typical twisted pair cabling (e.g., Category 5, Category 5e, Category 6, etc.). Due to the high bandwidth, low latency interconnections and network architecture, the data center 100 may, in use, pool resources, such as memory, accelerators (e.g., graphics accelerators, FPGAs, application-specific integrated circuits (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-manipulatable resource sleds. Furthermore, in the illustrative embodiment, the racks 102A, 102B, 102C, 102D include integrated power sources that receive a greater voltage than is typical for power sources. The increased voltage enables the power sources to provide additional power to the components on each sled, enabling the components to operate at higher than typical frequencies.

FIG. 2 illustrates an exemplary logical configuration of a rack 202 of the data center 100. As shown in FIG. 2, rack 202 may generally house a plurality of sleds, each of which may comprise a respective set of physical resources. In the particular non-limiting example depicted in FIG. 2, rack 202 houses sleds 204-1 to 204-4 comprising respective sets of physical resources 205-1 to 205-4, each of which constitutes a portion of the collective set of physical resources 206 comprised in rack 202. With respect to FIG. 1, if rack 202 is representative of—for example—rack 102A, then physical resources 206 may correspond to the physical resources 105A comprised in rack 102A. In the context of this example, physical resources 105A may thus be made up of the respective sets of physical resources, including physical storage resources 205-1, physical accelerator resources 205-2, physical memory resources 205-3, and physical compute resources 205-5 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 manipulatable sleds comprising disaggregated resources, each type of resource can be upgraded independently of each other and at their own optimized refresh rate.

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

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

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

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

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

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

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

FIG. 9 illustrates an example of a rack 902 that may be representative of a rack implemented according to rack architecture 800 of FIG. 8 according to some embodiments. In the particular non-limiting example depicted in FIG. 9, rack 902 features seven sled spaces 903-1 to 903-7, which include respective primary regions 903-1A to 903-7A and respective expansion regions 903-1B to 903-7B. In various embodiments, temperature control in rack 902 may be implemented using an air cooling system. For example, as reflected in FIG. 9, rack 902 may feature a plurality of fans 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 heat pipe and/or heat sinks arranged to dissipate heat generated by physical resources 1005. It is worthy of note that although the example sled 1004 depicted in FIG. 10 does not feature an expansion connector, any given sled that features the design elements of sled 1004 may also feature an expansion connector according to some embodiments. The embodiments are not limited in this context.

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

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

In another example, in various embodiments, one or more pooled storage sleds 1132 may be included among the physical infrastructure 1100A of data center 1100, each of which may comprise a pool of storage resources that is 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 quality of service (QoS) management capabilities for cloud services 1140. The embodiments are not limited in this context.

Referring now to FIG. 12, an illustrative compute device 1200 for automatic discovery of fault domains includes a processor 1220, an input/output (I/O) subsystem 1222, a memory 1224, and a data storage device 1226. The compute device 1200 may be embodied as server computer, a rack server, a blade server, a compute node, and/or a sled in a data center, such as a sled 204 as described above in connection with FIG. 2, a sled of the physical infrastructure 1100A as described above in connection with FIG. 11, or another sled of the data center.

The processor 1220 may be embodied as any type of processor capable of performing the functions described herein. For example, the processor 1220 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 1224 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 1224 may store various data and software used during operation of the compute device 1200 such operating systems, applications, programs, libraries, and drivers. The memory 1224 is communicatively coupled to the processor 1220 via the I/O subsystem 1222, which may be embodied as circuitry and/or components to facilitate input/output operations with the processor 1220, the memory 1224, and other components of the compute device 1200. For example, the I/O subsystem 1222 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 1222 may form a portion of a system-on-a-chip (SoC) and be incorporated, along with the processor 1220, the memory 1224, and other components of the compute device 1200, on a single integrated circuit chip.

The data storage device 1226 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 compute device 1200 may also include a communications circuitry 1228, which may be embodied as any communication circuit, device, or collection thereof, capable of enabling communications between the compute device 1200 and other remote devices over a computer network (not shown). The communications circuitry 1228 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.

The compute device 1200 may further include one or more peripheral devices 1232. The peripheral devices 1232 may include any number of additional input/output devices, interface devices, hardware accelerators, and/or other peripheral devices. For example, in some embodiments, the peripheral devices 1232 may include a touch screen, graphics circuitry, a graphical processing unit (GPU) and/or processor graphics, an audio device, a microphone, a camera, a keyboard, a mouse, a network interface, and/or other input/output devices, interface devices, and/or peripheral devices.

Referring now to FIG. 13, the compute device 1200 may establish an environment 1300 during operation. The illustrative environment 1300 includes a fault domain manager 1302 and a distributed processing adapter 1304. Each of the components of the environment 1300 may be embodied as hardware, firmware, software, or a combination thereof. As such, in some embodiments, one or more of the components of the environment 1300 may be embodied as circuitry or a collection of electrical devices (e.g., fault domain manager 1302 circuitry and distributed processing adapter 1304 circuitry, etc.). It should be appreciated that, in such embodiments, one or more of the fault domain manager 1302 circuitry and distributed processing adapter 1304 circuitry may form a portion of one or more of the compute device 1200, the communication circuitry 1228, the I/O subsystem 1222, and/or other components of the compute device 1200.

Additionally, the illustrative environment 1300 includes fault domain data 1306 which may be embodied as any data indicative of rack data, datacenter manager computer data, network connectivity data, power source device data, power zone data, or the like. The illustrative environment 1300 further includes sled data 1308, which may be embodied as any data indicative of sled identification data, which may further include data regarding compute devices connected to a particular sled (e.g., a sled 1718 as described in greater detail below with respect to FIG. 17). For example, sled identification data may include identifiers for power source devices, rack compute devices, top-of-rack switch devices, and any other devices directly or indirectly connected to a sled. Sled identification data may further include data regarding sled capabilities, manufacturer and device type/class identifiers, or the like. Sled data may further include sled health data. In the illustrative embodiment, sled health data may be embodied as any data indicative of a number of sockets, a number of cores, a number of drives, a latency metric, a reliability metric, and a memory capacity metric associated with the sled. Sled health data may also be embodied as any data indicative of other metrics such as sled performance data, sled performance history data, or the like.

In addition, both sled identification data and sled health data may be embodied as any data indicative or particular qualities or capabilities of the sled. For example, sled data for a compute sled may be embodied as any data indicative of a number of sockets, a number of cores, speed, memory types, memory modes, memory capacity and memory performance, NIC data or the like for the compute sled. Similarly, sled data for a storage sled may be embodied as any data indicative of a number of drives, drive information (e.g. performance, latency, bandwidth}, capacity, endurance, reliability (e.g., replication metrics), security, or the like. Sled data for an accelerator sled may be embodied as any data indicative of a number of accelerator devices (e.g., ASICs, FPGAs), accelerator device performance, an accelerator gate count, identifiers related to bitstreams of data input or output from the accelerator device, network address data, health data, or the like. Sled data for a network sled may be embodied as any data indicative of a number of ports, per port performance, network configuration, port connectivity, health data, or the like.

In the illustrative environment 1300, the fault domain manager 1302, which may be embodied as hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof as discussed above, is configured to receive sled data for at least one sled in a computer network. For example, the fault domain manager 1302 may receive sled data for one of the sleds 1716 that are described in greater detail below with respect to FIG. 17. As described herein, sled data may include one or more of sled identification data and sled health data. In the illustrative embodiment, the fault domain manager 1302 receives sled data each time a sled boots up. For example, each sled may be configured to transmit sled data to a connected fault domain manager 1302 at boot time. In the illustrative embodiment, the sled employs the link layer discovery protocol (LLDP) to generate sled data information. As used herein, LLDP refers to a network protocol that can be used by a device to advertise its identity, capabilities, and neighbors. When activated, LLDP is used by a device to send messages (also referred as advertisements) that include data that may be sent in the form of an Ethernet frame. Each frame contains one LLDP Data Unit (LLDPDU). Each LLDPDU is a sequence of type-length-value (TLV) structures. In the illustrative embodiment, the LLDP is extended to enable a sled to provide sled data as described herein. Accordingly, the fault domain manager 1302 is configured to receive one or more LLDPUs, some of which will contain one or more items of sled data as described above.

The fault domain manager 1302 is also configured to parse the sled identification data to identify a power zone. A datacenter may be organized into one or more power zones, each of which may include a group of power source devices (e.g., main power lines, unlimited power supply devices, backup generator devices, or the like). Additionally, each power zone is associated with one or more sleds (e.g., power zone A 1710 as described below with respect to FIG. 17). One power zone may be associated with multiple sleds. In other words, power source devices grouped into one power zone may be supplying power services to multiple sleds. Accordingly, the fault domain manager 1302 identifies and stores data regarding the power zone associated with a sled. In the illustrative embodiment, the fault domain manager receives sled data for multiple sleds and parses sled identification data for each sled. The fault domain manager 1302 is also configured to collate or otherwise group sleds by power zone identifier. For example, the fault domain manager 1302 is configured to group all sleds that are associated with a particular power zone. The fault domain manager 1302 is configured to perform this discovery process for each sled for which sled data is available.

Using the collated sled data and power zone identifier data generated from the parsing, the fault domain manager 1302 is also configured to generate a fault domain mapping using the identified power zone. More specifically, the fault domain manager 1302 generates a fault domain mapping that includes the power zone. A fault domain may describe the structure or organization of the group of compute devices that are associated with a particular power zone. The fault domain may include rack devices, switch devices, power source devices, and sleds. Accordingly, the fault domain mapping may be embodied as any data indicative of interconnections between rack devices, switch devices, power source devices, and sleds. For example, the sled identification data may be embodied as any data indicative of the top-of-rack switches that a sled is connected to, as well as the rack device(s) that each of the identified top-of-rack switches is connected to.

The fault domain manager 1302 is also configured to store the generated fault domain mapping in a data structure. For example, the fault domain manager 1302 may store the fault domain mapping in a tree structure, with a power zone identifier represented by a root or parent node with rack devices represented as child nodes, switches represented as child nodes of the rack device nodes, and sleds represented as child nodes of the switch device nodes. As another example, the fault domain manager 1302 may store the fault domain mapping in a graph database, with each device stored as a node object and relationship data stored in edge objects (e.g., top-of-rack switch X connects to sled Y and sled Z).

The distributed processing adapter 1304, which may be embodied as hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof, is configured to convert the generated fault domain mapping into a consumable fault domain mapping that is consumable by a distributed processing software system (sometimes also referred to as a “distributed storage software system”). More specifically, the distributed processing adapter adapts the fault domain mapping data (e.g., that is stored in a data structure as described above) and programmatically converts into a particular format. The format may be determined based on the target distributed processing software system.

As used herein, “distributed processing software system” refers to any software or software framework used for distributed storage and processing of large datasets. Frequently, a distributed processing software system includes a storage component, one that may be embodied as a distributed file system, for example. Also, a distributed processing software system includes a processing component to process the data from the file system. For example, a programming model and/or schema may be provided that provides interfaces for object-, block- and file-level storage. Distributed processing software systems often enable processing of large datasets by distributing the large dataset into smaller chunks across multiple processing nodes, delivering packaged processing code to those nodes, and processing these chunks in parallel. One example of a distributed processing software system is Apache Hadoop® (APACHE and HADOOP are registered trademarks of the Apache Software Foundation, located at Forest Hill, Md., USA). As an example, a Hadoop distributed file system may be able to consume the fault domain mapping data if it is formatted for consumption by code written in the Java® programming language (JAVA is a registered trademark of the Oracle Corporation, located at Redwood Shores, Redwood City, Calif., USA). Accordingly, the distributed processing adapter 1304 formats the fault domain mapping data into a format consumable by a Hadoop architecture. After conversion, the distributed processing adapter 1304 provides the fault domain mapping to the distributed processing software system.

Another example of a distributed processing software system is Ceph (CEPH is a registered trademark of Red Hat Inc., located at Raleigh, N.C., USA). Yet another example of a distributed processing software system is Cassandra (CASSANDRA is a registered trademark of Apache Software Foundation, located at Forest Hill, Md., USA). Yet another example of a distributed processing software system is MongoDB (MONGODB is a registered trademark of MongoDB Inc., located at Palo Alto, Calif., United States).

It should be appreciated that each of the fault domain manager 1302 and the distributed processing adapter 1304 may be separately embodied as hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof. For example, the distributed processing adapter 1304 may be embodied as a hardware component, while the fault domain manager 1302 is embodied as a virtualized hardware component or as some other combination of hardware, firmware, software, virtualized hardware, emulated architecture, and/or a combination thereof. Further it should be appreciated that in some embodiments, a sled 1718 (as described in greater detail below with respect to FIG. 17) may establish an environment similar to the environment 1300 described above.

Referring now to FIG. 14, in use, a sled 1718 may execute a method 1400 for automatic discovery of fault domains. More specifically, FIG. 14 describes the processes a sled 1718 executes to generate sled data that will later be used by, for example, the fault domain manager 1302. The method 1400 begins in block 1402, in which the sled 1718 initiates a sled boot process. In the illustrative embodiment, the sled 1718 is configured to, for example, initiate loading of operating system program loading, as in block 1404.

In block 1406, the sled 1718 performs a sled data discovery process. In the illustrative embodiment, sled data discovery is performed using an LLDP protocol extension. As described above, LLDP is used to transmit advertisements by a device including data regarding the device's identity, capabilities, neighbors, and the like. Accordingly, in block 1408, the sled 1718 discovers its capabilities (for example, those described above with respect to sled identification data and sled health data). In block 1410, the sled 1718 discovers the sled 1718's position (e.g., with respect to other sleds and other devices). For example, the sled 1718 may be a member of a group of compute sleds, accelerator sleds, or other sleds loaded in a single rack. In addition to determining its neighbors, in block 1412, the sled 1718 discovers the rack device(s), switch device(s), and power source devices that it is directly or indirectly connected to. In one embodiment, the sled 1718 queries connected devices with specific queries regarding rack, switch, and power source devices that are associated with the sled 1718. In other embodiments, the sled 1718 itself may receive, for example, LLDP advertisements that bear identification data for other devices. For example, a rack device may send the sled 1718 an LLDP advertisement indicating to the sled 1718 that the sled 1718 is connected to that rack device.

Relatedly, in block 1414, the sled 1718 is configured to perform discovery protocols (e.g., LLDP) across switch domains. For example, the sled 1718 may be configured to receive advertisements from specific devices. Additionally, a compute device may be configured during installation to transmit data to a specific sled (or any compute device). Using the example of LLDP, the LLDP protocol (which may be disabled by default) will be activated on a specific compute device. As an example, the compute device (e.g., a rack device) may be configured during installation to transmit periodic LLDP advertisements to all connected switch devices, which are, in turn, configured during installation to transmit periodic LLDP advertisements to all connected sleds. Referring now to FIG. 17, a diagram 1700 illustrates a potential connectivity between rack devices, switch devices, and sleds. This connectivity may be advertised to a particular sled 1718 using the methods illustrated above.

In block 1416, the sled 1718 transmits the discovered sled data to a fault domain manager 1302. It should be appreciated that, in some embodiments, the fault domain manager 1302 may be a sub-component of the sled 1718 that is executing method 1400. As a result, the generated sled data will be passed to the fault domain manager 1302 of the sled 1718. In other embodiments, the fault domain manager 1302 may be a component of another sled or compute device, in which case the sled 1718 transmits the generated sled data to the remote fault domain manager 1302 located on another compute device.

Referring now to FIG. 15, in use, the compute device 1200 may execute a method 1500 for automatic discovery of fault domains. It should be appreciated that, in some embodiments, the operations of the method 1500 may be performed by one or more components of the compute device 1200 as shown in FIG. 13, such as a processor 1220. More specifically, the processor 1220 may execute (in hardware or software) at least one of the fault domain manager 1302 and the distributed processing adapter 1304. The method 1500 may also be performed by one or more components of a sled 1718 as shown in FIG. 17.

The method 1500 begins in block 1502, in which the compute device 1200 checks for sled boot up. As described above, method 1500 may be performed by a sled 1718 which in turn detects whether it has been booted up. If method 1500 is performed by another compute device 1200, the compute device 1200 is configured to check monitoring applications that monitor sled activity and transmit alerts to the compute device 1200, as shown in block 1504. In block 1506, the compute device 1200 receives sled boot up confirmation (e.g., directly from a sled 1718 or from a monitoring application). In block 1508, the compute device 1200 receives sled data from a sled 1718. Referring now to FIG. 17, a diagram 1700 illustrates one potential computing network in which multiple sleds are interconnected with various other types of devices. Any one of the devices illustrated in FIG. 17 may represent the compute device 1200 and may perform method 1500 and receive sled data for one or more of sleds 1718.

Referring back to FIG. 15, in block 1510, the compute device 1200 receives sled identification data (as described above with respect to sled data 1308). Similarly, in block 1512, the compute device 1200 receives sled health data (as described above with respect to sled data 1308). In block 1514, the compute device 1200 parses sled data to identify at least one power zone corresponding to the sled 1718. More specifically, the compute device 1200 identifies at least one rack device, switch device, and/or power source device connected directly or indirectly to the sled 1718. If the sled data is transmitted, for example, as an LLDP advertisement, the compute device 1200 is configured to parse the sled data by reading each LLDPU and extracting one or more type-length-value structures (TLVs) inside the LLDPU that represent one or more items of sled data. For example, one TLV may be (2: 9 bits: 1234), where type 2 represents the data type (e.g., Port ID), length 9 bits represents the length of the value, and 1234 represents the value itself (e.g., a port identifier for a port associated with a sled 1718).

The compute device 1200 is also configured to parse specific types of incoming sled data. In the illustrative embodiment, in block 1516, the compute device 1200 parses storage sled data, which may be embodied as any data indicative of a number of drives, drive information (e.g. performance, latency, bandwidth), capacity, endurance, reliability (e.g., replication metrics), security, or the like. In block 1518, the compute device 1200 parses compute sled data, which may be embodied as any data indicative of a number of sockets, a number of cores, speed, memory types, memory modes, memory capacity and memory performance, NIC data or the like. In block 1520, the compute device 1200 parses accelerator sled data, which may be embodied as any data indicative of a number of accelerator devices (e.g., ASICs, FPGAs), accelerator device performance, an accelerator gate count, identifiers related to bitstreams of data input or output from the accelerator device, network address data, health data, or the like.

In block 1522, the compute device 1200 identifies a rack device associated with the sled 1718 based on the sled identification data. In the illustrative embodiment, the compute device 1200 may determine the rack device from sled data received directly from the sled 1718. In another embodiment, the compute device 1200 may determine the rack device from sled data received from another device connected to the sled 1718 (e.g., a top-of-rack switch device). In some embodiments, for example, the compute device 1200 uses an LLDP extension to enable the native LLDP protocol to include an optional TLV that bears data regarding the rack device associated with the sled 1718.

In block 1524, the compute device 1200 identifies a manager computer associated with the sled 1718 based on sled identification data. In the illustrative embodiment, a manager computer is any compute device (e.g., an orchestrator server or resource manager computer) that manages one or more power zones (or one or more rack devices in conjunction with other devices). In the illustrative embodiment, the compute device 1200 may determine the manager computer from sled data received directly from the sled 1718. In another embodiment, the compute device 1200 may determine the manager computer from sled data received from another device connected to the sled 1718 (e.g., a top-of-rack switch device). In the illustrative embodiment, for example, the compute device 1200 uses an LLDP extension to enable the native LLDP protocol to include an optional TLV that bears data regarding the manager computer associated with the sled 1718.

The method 1500 continues at block 1526, where the compute device 1200 determines a switch that is associated with a rack device for a rack which the sled 1718 is located or otherwise associated with the sled 1718. In one embodiment, the sled 1718 may be part of a group of sleds that are located within a rack where the sleds are connected to computers outside the rack using a top-of-rack switch device 1716. In block 1528, the compute device 1200 identifies power source device based on the determined top-of-rack switch device 1716. In block 1530, the compute device 1200 determines whether there are additional sleds for which sled data is available. If there are more sleds for which sled data is available, the method returns to block 1508 to receive more sled data. If there are no more sleds for which sled data is available, the method advances to block 1532, shown in FIG. 16.

Referring now to FIG. 17, a diagram 1700 illustrates one potential computing network in which multiple sleds are interconnected with various other types of devices. As shown, a logical root is connected to one or more datacenters 1704. More specifically, the root 1702 may be a system operator's computer, a mainframe computer used to manage a large number of compute devices, or the like. Each datacenter 1704 may also be thought of as one or more manager computers as described above with respect to block 1524. A datacenter 1704 is connected to one or more power zones 1706, 1708, 1710, 1712. As described above, each power zone represents a collection of power source devices (e.g., main power lines, unlimited power supply devices, backup generator devices, or the like) that is in turn connected to one or more rack devices 1714. Each rack device is connected to one or more switches 1716, represented herein as top-of-rack switches 1716. Each top-of-rack switch 1716 is connected to one or more sleds 1718 (which may be placed inside or associated with a rack device 1714). FIG. 17 also illustrates a fault domain 1722. The fault domain 1722 may be embodied as any data indicative of a subset of the computing network. The fault domain 1722 may be defined as the group of devices that rely on the same power zone 1706 (which may include various power source devices). This reliance may be represented by an understanding that, were one or more of the power source devices in power zone 1706 to fail or experience a fault, the issue would be likely to affect one or more other devices in the fault domain 1722 (e.g., racks, switches, or sleds) but would leave devices outside the fault domain 1722 unaffected. Moreover, a composed node 1720 is shown as a combination of one or more sleds 1718, where each sled may perform specific functions. In the case of a composed node, sleds 1718 may be selected across fault domains. For example, one sled in the composition may be a compute sled from one fault domain, another an accelerator sled from another fault domain, and so on. As a result, the illustrative embodiment provides a virtual-to-physical topology that represents the connectivity and, relatedly, the power zone for a composed node 1720.

Referring now to FIG. 16, in use, the compute device 1200 continues to execute the method 1500 for automatic discovery of fault domains. The method 1500 continues at block 1532, in which the compute device 1200 generates a fault domain mapping based on the sled data. More specifically, and referring to block 1534, the compute device 1200 generates the fault domain mapping based on switch data. For example, switch data for each sled 1718 is collated by sled and the sleds connected to each switch (e.g., a top-of-rack switch in a rack) are stored in a data structure (e.g., a tree or graph) in association with an identifier for the switch. In block 1536, the compute device 1200 generates the fault domain mapping based on power source device data. Power source device data (or power zone data) for each sled 1718 is collated by sled and an identifier for a power source device (e.g., an AC power source) is stored in a data structure (e.g., a tree or graph) in association with each connected sled. In block 1538, the compute device 1200 generates the fault domain mapping based on rack data. For example, an identifier for the rack device 1714 that holds the sled is stored in a data structure (e.g., a tree or graph) in association with identifiers for the sled. In block 1540, the compute device 1200 generates the fault domain mapping based on manager computer data. For example, an identifier for the manager computer of a datacenter 1704 that manages the sled is stored in a data structure (e.g., a tree or graph) in association with identifiers for the sled.

In block 1542, the compute device 1200 collates the power source device identifiers to determine the group of sleds associated with the same power zone. In the illustrative embodiment, each power source device may supply power across rack domains and across switch domains. Grouping sleds by power zone may generate a fault domain 1722, as illustrated in FIG. 17. The compute device proceeds to generate a fault domain for each sled and complete a fault domain mapping that includes each fault domain.

In block 1544, the compute device 1200 may programmatically convert the fault domain mapping into a consumable fault domain mapping. More specifically, the consumable fault domain mapping is configured to be consumed by a distributed data processing software system, as described above with respect to FIG. 13. In the illustrative embodiment, as shown in block 1546, the compute device 1200 stores the software performing the conversion in the form of a plug-in application. When a particular distributed data processing software system is the target for the consumable fault domain mapping, the compute device 1200 selects the plug-in application that corresponds to the targeted distributed data processing software system and uses that plug-in application to perform the conversion. For example, the plug-in applications for each distributed data processing software system may be components of the distributed processing adapter 1304. Thus, the fault domain discovery process of the compute device 1200 may be adapted to multiple data processing software systems.

Relatedly, in block 1548, the compute device 1200 converts a fault domain mapping for a composed node. As described above with respect to FIG. 17 sleds 1718 may be selected across fault domains in the case of a composed node. Accordingly, the consumable fault domain mapping reflects the cross-domain sled composition of the composed node. In block 1550, the compute device 1200 transmits the converted fault domain mapping to distributed data processing software. In a related embodiment, as shown in block 1552, the compute device 1200 may expose the converted fault domain mapping to distributed data processing software using an API. The distributed data processing software may use the fault domain mapping information, for example, to distribute application data to multiple fault domains or to otherwise improve fault tolerance. After providing the fault domain mapping to the data processing software system, the method 1500 loops back to block 1502, shown in FIG. 15, to continue generating fault domain mappings.

EXAMPLES

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

Example 1 includes a compute device to auto-discover power system fault domains within a computer network, the compute device 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 compute device to: receive sled data for at least one sled of a plurality of sleds in the computer network, the sled data including sled identification data and sled health data; parse the sled identification data to identify at least one power zone wherein each power zone includes a subset of the plurality of sleds; generate a fault domain mapping using the at least one identified power zone; convert the generated fault domain mapping into a consumable fault domain mapping that is consumable by a distributed processing software system; and provide the consumable fault domain mapping to the distributed processing software system.

Example 2 includes the subject matter of Example 1, and wherein to receive the sled data comprises to receive an advertisement generated by the at least one sled via a link layer discovery protocol (LLDP), wherein LLDP is configured to identify one or more relationships between one or more sleds of the plurality of sleds, and wherein the sled health data includes one or more of a number of sockets, a number of cores, a number of drives, a latency metric, a reliability metric, and a memory capacity metric.

Example 3 includes the subject matter of any of Examples 1 and 2, 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 compute device to identify, within the sled identification data, a first top-of-rack switch identifier for a first top-of-rack switch that is communicatively coupled to the at least one sled of the plurality of sleds.

Example 4 includes the subject matter of any of Examples 1-3, 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 compute device to identify, within the sled identification data, a second top-of-rack switch identifier for a second top-of-rack switch that is distinct from the first top-of-rack switch, wherein the second-top-of-rack switch is communicatively coupled to the at least one sled.

Example 5 includes the subject matter of any of Examples 1-4, 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 compute device to determine that the first top-of-rack switch and the second top-of-rack switch are communicatively coupled to a rack compute device.

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 compute device to determine that the rack compute device is communicatively coupled to at least one power source device, wherein the at least one power source device is associated with the at least one power zone.

Example 7 includes the subject matter of any of Examples 1-6, 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 compute device to determine that the rack compute device is communicatively coupled to at least one other power source device, wherein the at least one other power source device is associated with at least one other power zone.

Example 8 includes the subject matter of any of Examples 1-7, and wherein the computer network includes a composed node, wherein the composed node includes the at least one sled, 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 compute device to: determine that the at least one sled corresponds to the at least one identified power zone; and identify that the composed node corresponds to the at least one identified power zone within the fault domain mapping.

Example 9 includes the subject matter of any of Examples 1-8, 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 compute device to identify, within the sled identification data, a datacenter identifier for a datacenter, wherein the datacenter includes at least one manager computer that manages at least one sled of the plurality of sleds, and wherein the manager computer is communicatively coupled to the at least one sled.

Example 10 includes the subject matter of any of Examples 1-9, 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 compute device to: determine a data format corresponding to the distributed processing software system; select a plug-in application based on the data format; and convert the generated fault domain mapping into a consumable fault domain mapping by executing the selected plug-in application.

Example 11 includes the subject matter of any of Examples 1-10, and wherein the distributed processing software system includes at least one of Apache Hadoop, Apache Cassandra, Ceph, and MongoDB.

Example 12 includes a compute sled to transmit sled data to a fault domain manager within a computer network, the compute 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 compute sled to: initiate a boot process for the compute sled; discover sled data, wherein the sled data includes sled identification data and sled health data, wherein to discover the sled identification data comprises to identify a first top-of-rack switch, wherein the first top-of-rack switch is communicatively coupled to the compute sled; generate an advertisement that includes the sled data, wherein the advertisement is configured to identify at least one relationship between the compute sled and at least one other compute sled within the computer network; and transmit the sled data to the fault domain manager.

Example 13 includes the subject matter of Example 12, wherein to initiate the boot process comprises to initiate an operating system program load process.

Example 14 includes the subject matter of any of Examples 12 and 13, wherein to discover the sled data includes to discover, within the sled identification data, a second top-of-rack switch identifier for a second top-of-rack switch that is distinct from the first top-of-rack switch, wherein the second-top-of-rack switch is communicatively coupled to the compute sled.

Example 15 includes the subject matter of any of Examples 12-14, wherein the first top-of-rack switch and the second top-of-rack switch are communicatively coupled to a rack compute device.

Example 16 includes the subject matter of any of Examples 12-15, wherein the rack compute device is communicatively coupled to at least one power source device, wherein the at least one power source device is associated with at least one power zone.

Example 17 includes the subject matter of any of Examples 12-16, wherein the rack compute device is communicatively coupled to at least one other power source device, wherein the at least one other power source device is associated with at least one other power zone.

Example 18 includes the subject matter of any of Examples 12-17, wherein to discover the sled data includes to discover, within the sled identification data, sled identification data further includes a datacenter identifier for a datacenter, wherein the datacenter includes at least one manager computer that manages at least one sled of the plurality of sleds, and wherein the manager computer is communicatively coupled to the at least one sled.

Example 19 includes the subject matter of any of Examples 12-18, wherein the sled health data includes one or more of a number of sockets, a number of cores, a number of drives, a latency metric, a reliability metric, and a memory capacity metric.

Example 20 includes the subject matter of any of Examples 12-19, wherein to generate the advertisement comprises to generate the advertisement using a link layer discovery protocol (LLDP).

Example 21 includes a method of automatically discovering power system fault domains within a computer network, the method comprising: receiving sled data for at least one sled of a plurality of sleds in the computer network, the sled data including sled identification data and sled health data; parsing the sled identification data to identify at least one power zone wherein each power zone includes a subset of the plurality of sleds; generating a fault domain mapping using the at least one identified power zone; converting the generated fault domain mapping into a consumable fault domain mapping that is consumable by a distributed processing software system; and providing the consumable fault domain mapping to the distributed processing software system.

Example 22 includes the subject matter of Example 21, and further comprising receiving sled data by an advertisement generated by the at least one sled via a link layer discovery protocol (LLDP), wherein LLDP is configured to identify one or more relationships between one or more sleds of the plurality of sleds, and wherein the sled health data includes one or more of a number of sockets, a number of cores, a number of drives, a latency metric, a reliability metric, and a memory capacity metric.

Example 23 includes the subject matter of any of Examples 21 and 22, and further comprising identifying, within the sled identification data, a first top-of-rack switch identifier for a first top-of-rack switch that is communicatively coupled to the at least one sled of the plurality of sleds.

Example 24 includes the subject matter of any of Examples 21-23, and further comprising identifying, within the sled identification data, a second top-of-rack switch identifier for a second top-of-rack switch that is distinct from the first top-of-rack switch, wherein the second-top-of-rack switch is communicatively coupled to the at least one sled.

Example 25 includes the subject matter of any of Examples 21-24, and further comprising determining that the first top-of-rack switch and the second top-of-rack switch are communicatively coupled to a rack compute device.

Example 26 includes the subject matter of any of Examples 21-25, and further comprising determining that the rack compute device is communicatively coupled to at least one power source device, wherein the at least one power source device is associated with the at least one power zone.

Example 27 includes the subject matter of any of Examples 21-26, and further comprising determining that the rack compute device is communicatively coupled to at least one other power source device, wherein the at least one other power source device is associated with at least one other power zone.

Example 28 includes the subject matter of any of Examples 21-27, and wherein the computer network includes a composed node, wherein the composed node includes the at least one sled, and further comprising: determining that the at least one sled corresponds to the at least one identified power zone; and identifying that the composed node corresponds to the at least one identified power zone within the fault domain mapping.

Example 29 includes the subject matter of any of Examples 21-28, and further comprising identifying, within the sled identification data, a datacenter identifier for a datacenter, wherein the datacenter includes at least one manager computer that manages at least one sled of the plurality of sleds, and wherein the manager computer is communicatively coupled to the at least one sled.

Example 30 includes the subject matter of any of Examples 21-29, and further comprising: determining a data format corresponding to the distributed processing software system; selecting a plug-in application based on the data format; and converting the generated fault domain mapping into a consumable fault domain mapping by executing the selected plug-in application.

Example 31 includes the subject matter of any of Examples 21-30, and wherein the distributed processing software system includes at least one of Apache Hadoop, Apache Cassandra, Ceph, and MongoDB.

Example 32 includes a method of transmitting sled data to a fault domain manager within a computer network, the method comprising: initiating, by a compute sled, a boot process for the compute sled; discovering, by the compute sled, sled data, wherein the sled data includes sled identification data and sled health data, wherein to discover the sled identification data comprises to identify a first top-of-rack switch, wherein the first top-of-rack switch is communicatively coupled to the compute sled; generating, by the compute sled, an advertisement that includes the sled data, wherein the advertisement is configured to identify at least one relationship between the compute sled and at least one other compute sled within the computer network; and transmitting, by the compute sled, the sled data to the fault domain manager.

Example 33 includes the subject matter of Example 32, and wherein initiating the boot process comprises initiating initiate an operating system program load process.

Example 34 includes the subject matter of any of Examples 32 and 33, and wherein discovering the sled data includes discovering, within the sled identification data, a second top-of-rack switch identifier for a second top-of-rack switch that is distinct from the first top-of-rack switch, wherein the second-top-of-rack switch is communicatively coupled to the compute sled.

Example 35 includes the subject matter of any of Examples 32-34, and wherein the first top-of-rack switch and the second top-of-rack switch are communicatively coupled to a rack compute device.

Example 36 includes the subject matter of any of Examples 32-35, and wherein the rack compute device is communicatively coupled to at least one power source device, wherein the at least one power source device is associated with at least one power zone.

Example 37 includes the subject matter of any of Examples 32-36, and wherein the rack compute device is communicatively coupled to at least one other power source device, wherein the at least one other power source device is associated with at least one other power zone.

Example 38 includes the subject matter of any of Examples 32-37, and wherein discovering the sled data includes discovering, within the sled identification data, sled identification data further includes a datacenter identifier for a datacenter, wherein the datacenter includes at least one manager computer that manages at least one sled of the plurality of sleds, and wherein the manager computer is communicatively coupled to the at least one sled.

Example 39 includes the subject matter of any of Examples 32-38, and wherein the sled health data includes one or more of a number of sockets, a number of cores, a number of drives, a latency metric, a reliability metric, and a memory capacity metric.

Example 40 includes the subject matter of any of Examples 32-39, and wherein generating the advertisement comprises generating the advertisement using a link layer discovery protocol (LLDP).

Example 41 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 21-40.

Example 42 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 21-40.

Example 43 includes a computing device comprising means for performing the method of any of Examples 21-40.

Example 44 includes a compute device to auto-discover power system fault domains within a computer network, the compute device comprising: fault domain manager circuitry to: receive sled data for at least one sled of a plurality of sleds in the computer network, the sled data including sled identification data and sled health data; parse the sled identification data to identify at least one power zone wherein each power zone includes a subset of the plurality of sleds; generate a fault domain mapping using the at least one identified power zone; and distributed processing adapter circuitry to: convert the generated fault domain mapping into a consumable fault domain mapping that is consumable by a distributed processing software system; and provide the consumable fault domain mapping to the distributed processing software system.

Example 45 includes the subject matter of Example 44, and wherein to receive the sled data comprises to receive an advertisement generated by the at least one sled via a link layer discovery protocol (LLDP), wherein LLDP is configured to identify one or more relationships between one or more sleds of the plurality of sleds, and wherein the sled health data includes one or more of a number of sockets, a number of cores, a number of drives, a latency metric, a reliability metric, and a memory capacity metric.

Example 46 includes the subject matter of any of Examples 44 and 45, and wherein the fault domain manager circuitry is further to identify, within the sled identification data, a first top-of-rack switch identifier for a first top-of-rack switch that is communicatively coupled to the at least one sled of the plurality of sleds.

Example 47 includes the subject matter of any of Examples 44-46, and wherein the fault domain manager circuitry is further to identify, within the sled identification data, a second top-of-rack switch identifier for a second top-of-rack switch that is distinct from the first top-of-rack switch, wherein the second-top-of-rack switch is communicatively coupled to the at least one sled.

Example 48 includes the subject matter of any of Examples 44-47, and wherein the fault domain manager circuitry is further to determine that the first top-of-rack switch and the second top-of-rack switch are communicatively coupled to a rack compute device.

Example 49 includes the subject matter of any of Examples 44-48, and wherein the fault domain manager circuitry is further to determine that the rack compute device is communicatively coupled to at least one power source device, wherein the at least one power source device is associated with the at least one power zone.

Example 50 includes the subject matter of any of Examples 44-49, and wherein the fault domain manager circuitry is further to determine that the rack compute device is communicatively coupled to at least one other power source device, wherein the at least one other power source device is associated with at least one other power zone.

Example 51 includes the subject matter of any of Examples 44-50, and wherein the computer network includes a composed node, wherein the composed node includes the at least one sled, and wherein the fault domain manager circuitry is further to: determine that the at least one sled corresponds to the at least one identified power zone; and identify that the composed node corresponds to the at least one identified power zone within the fault domain mapping.

Example 52 includes the subject matter of any of Examples 44-51, and wherein the fault domain manager circuitry is further to identify, within the sled identification data, a datacenter identifier for a datacenter, wherein the datacenter includes at least one manager computer that manages at least one sled of the plurality of sleds, and wherein the manager computer is communicatively coupled to the at least one sled.

Example 53 includes the subject matter of any of Examples 44-52, and wherein the distributed processing adapter circuitry is further to: determine a data format corresponding to the distributed processing software system; select a plug-in application based on the data format; and convert the generated fault domain mapping into a consumable fault domain mapping by executing the selected plug-in application.

Example 54 includes the subject matter of any of Examples 44-53, and wherein the distributed processing software system includes at least one of Apache Hadoop, Apache Cassandra, Ceph, and MongoDB.

Example 55 includes a compute sled to transmit sled data to a fault domain manager within a computer network, the compute sled comprising: compute engine circuitry to: initiate a boot process for the compute sled; discover sled data, wherein the sled data includes sled identification data and sled health data, wherein to discover the sled identification data comprises to identify a first top-of-rack switch, wherein the first top-of-rack switch is communicatively coupled to the compute sled; generate an advertisement that includes the sled data, wherein the advertisement is configured to identify at least one relationship between the compute sled and at least one other compute sled within the computer network; and transmit the sled data to the fault domain manager.

Example 56 includes the subject matter of Example 55, wherein to initiate the boot process comprises to initiate an operating system program load process.

Example 57 includes the subject matter of any of Examples 55 and 56, wherein to discover the sled data includes to discover, within the sled identification data, a second top-of-rack switch identifier for a second top-of-rack switch that is distinct from the first top-of-rack switch, wherein the second-top-of-rack switch is communicatively coupled to the compute sled.

Example 58 includes the subject matter of any of Examples 55-57, wherein the first top-of-rack switch and the second top-of-rack switch are communicatively coupled to a rack compute device.

Example 59 includes the subject matter of any of Examples 55-58, wherein the rack compute device is communicatively coupled to at least one power source device, wherein the at least one power source device is associated with at least one power zone.

Example 60 includes the subject matter of any of Examples 55-59, wherein the rack compute device is communicatively coupled to at least one other power source device, wherein the at least one other power source device is associated with at least one other power zone.

Example 61 includes the subject matter of any of Examples 55-60, wherein to discover the sled data includes to discover, within the sled identification data, sled identification data further includes a datacenter identifier for a datacenter, wherein the datacenter includes at least one manager computer that manages at least one sled of the plurality of sleds, and wherein the manager computer is communicatively coupled to the at least one sled.

Example 62 includes the subject matter of any of Examples 55-61, wherein the sled health data includes one or more of a number of sockets, a number of cores, a number of drives, a latency metric, a reliability metric, and a memory capacity metric.

Example 63 includes the subject matter of any of Examples 55-62, wherein to generate the advertisement comprises to generate the advertisement using a link layer discovery protocol (LLDP).

Example 64 includes a compute device of automatically discovering power system fault domains within a computer network, the compute device comprising: circuitry for receiving sled data for at least one sled of a plurality of sleds in the computer network, the sled data including sled identification data and sled health data; means for parsing the sled identification data to identify at least one power zone wherein each power zone includes a subset of the plurality of sleds; means for generating a fault domain mapping using the at least one identified power zone; means for converting the generated fault domain mapping into a consumable fault domain mapping that is consumable by a distributed processing software system; and means for providing the consumable fault domain mapping to the distributed processing software system.

Example 65 includes the subject matter of Example 64, and further comprising means for receiving sled data by an advertisement generated by the at least one sled via a link layer discovery protocol (LLDP), wherein LLDP is configured to identify one or more relationships between one or more sleds of the plurality of sleds, and wherein the sled health data includes one or more of a number of sockets, a number of cores, a number of drives, a latency metric, a reliability metric, and a memory capacity metric.

Example 66 includes the subject matter of any of Examples 64 and 65, and further comprising means for identifying, within the sled identification data, a first top-of-rack switch identifier for a first top-of-rack switch that is communicatively coupled to the at least one sled of the plurality of sleds.

Example 67 includes the subject matter of any of Examples 64-66, and further comprising means for identifying, within the sled identification data, a second top-of-rack switch identifier for a second top-of-rack switch that is distinct from the first top-of-rack switch, wherein the second-top-of-rack switch is communicatively coupled to the at least one sled.

Example 68 includes the subject matter of any of Examples 64-67, and further comprising means for determining that the first top-of-rack switch and the second top-of-rack switch are communicatively coupled to a rack compute device.

Example 69 includes the subject matter of any of Examples 64-68, and further comprising means for determining that the rack compute device is communicatively coupled to at least one power source device, wherein the at least one power source device is associated with the at least one power zone.

Example 70 includes the subject matter of any of Examples 64-69, and further comprising means for determining that the rack compute device is communicatively coupled to at least one other power source device, wherein the at least one other power source device is associated with at least one other power zone.

Example 71 includes the subject matter of any of Examples 64-70, and wherein the computer network includes a composed node, wherein the composed node includes the at least one sled, and further comprising: means for determining that the at least one sled corresponds to the at least one identified power zone; and means for identifying that the composed node corresponds to the at least one identified power zone within the fault domain mapping.

Example 72 includes the subject matter of any of Examples 64-71, and further comprising means for identifying, within the sled identification data, a datacenter identifier for a datacenter, wherein the datacenter includes at least one manager computer that manages at least one sled of the plurality of sleds, and wherein the manager computer is communicatively coupled to the at least one sled.

Example 73 includes the subject matter of any of Examples 64-72, and further comprising: means for determining a data format corresponding to the distributed processing software system; means for selecting a plug-in application based on the data format; and means for converting the generated fault domain mapping into a consumable fault domain mapping by executing the selected plug-in application.

Example 74 includes the subject matter of any of Examples 64-73, and wherein the distributed processing software system includes at least one of Apache Hadoop, Apache Cassandra, Ceph, and MongoDB.

Example 75 includes a compute device of transmitting sled data to a fault domain manager within a computer network, the compute device comprising: circuitry for initiating, by a compute sled, a boot process for the compute sled; means for discovering, by the compute sled, sled data, wherein the sled data includes sled identification data and sled health data, wherein to discover the sled identification data comprises to identify a first top-of-rack switch, wherein the first top-of-rack switch is communicatively coupled to the compute sled; means for generating, by the compute sled, an advertisement that includes the sled data, wherein the advertisement is configured to identify at least one relationship between the compute sled and at least one other compute sled within the computer network; and circuitry for transmitting, by the compute sled, the sled data to the fault domain manager.

Example 76 includes the subject matter of Example 75, and wherein the circuitry for initiating the boot process comprises circuitry for initiating initiate an operating system program load process.

Example 77 includes the subject matter of any of Examples 75 and 76, and wherein the means for discovering the sled data includes means for discovering, within the sled identification data, a second top-of-rack switch identifier for a second top-of-rack switch that is distinct from the first top-of-rack switch, wherein the second-top-of-rack switch is communicatively coupled to the compute sled.

Example 78 includes the subject matter of any of Examples 75-77, and wherein the first top-of-rack switch and the second top-of-rack switch are communicatively coupled to a rack compute device.

Example 79 includes the subject matter of any of Examples 75-78, and wherein the rack compute device is communicatively coupled to at least one power source device, wherein the at least one power source device is associated with at least one power zone.

Example 80 includes the subject matter of any of Examples 75-79, and wherein the rack compute device is communicatively coupled to at least one other power source device, wherein the at least one other power source device is associated with at least one other power zone.

Example 81 includes the subject matter of any of Examples 75-80, and wherein the means for discovering the sled data includes means for discovering, within the sled identification data, sled identification data further includes a datacenter identifier for a datacenter, wherein the datacenter includes at least one manager computer that manages at least one sled of the plurality of sleds, and wherein the manager computer is communicatively coupled to the at least one sled.

Example 82 includes the subject matter of any of Examples 75-81, and wherein the sled health data includes one or more of a number of sockets, a number of cores, a number of drives, a latency metric, a reliability metric, and a memory capacity metric.

Example 83 includes the subject matter of any of Examples 75-82, and wherein the means for generating the advertisement comprises means for generating the advertisement using a link layer discovery protocol (LLDP).

TECHNOLOGIES FOR AUTO-DISCOVERY OF FAULT DOMAINS

Information

Publication Number

Date Filed

Date Published

Inventors

Original Assignees

CPC

International Classifications

Abstract

Description

Claims

Priority Claims (1)

CROSS-REFERENCE TO RELATED APPLICATIONS

Provisional Applications (1)