MODULAR I/O CONFIGURATIONS FOR EDGE COMPUTING USING DISAGGREGATED CHIPLETS

• US 20200136906A1
• Filed: 12/20/2019
• Published: 04/30/2020
• Est. Priority Date: 04/30/2019
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Guim Bernat, Francesc, Doshi, Kshitij Arun, Smith, Ned M., Verrall, Timothy, Qureshi, Uzair
• Granted Patent: US 11,388,054 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 1/206 comprising thermal management
• CPC Class Codes: G06F 11/3006 where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems multiprogramming arrangements G06F9/46; allocation of resources G06F9/50
• CPC Class Codes: G06F 2209/5021 Priority
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
• CPC Class Codes: G06F 9/505 considering the load
• CPC Class Codes: G06F 9/5061 Partitioning or combining of resources
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: G06F 9/5094 where the allocation takes into account power or heat criteria power management in computers in general G06F1/3203; thermal management in computers in general G06F1/206
• CPC Class Codes: G06F 9/542 Event management; Broadcasting; Multicasting; Notifications
• CPC Class Codes: H04L 41/0843 based on generic templates
• CPC Class Codes: H04L 41/0869 Validating the configuration within one network element
• CPC Class Codes: H04L 41/5054 Automatic deployment of services triggered by the service manager, e.g. service implementation by automatic configuration of network components
• CPC Class Codes: H04L 47/781 Centralised allocation of resources
• CPC Class Codes: H04L 47/83 based on usage prediction
• CPC Class Codes: H04L 49/70 Virtual switches
• CPC Class Codes: H04L 63/00 Network architectures or network communication protocols for network security cryptographic mechanisms or cryptographic arrangements for secret or secure communication H04L9/00; network architectures or network communication protocols for wireless network security H04W12/00; security arrangements for protecting computers or computer systems against unauthorised activity G06F21/00
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 9/0637 Modes of operation, e.g. cipher block chaining [CBC], electronic codebook [ECB] or Galois/counter mode [GCM]
• CPC Class Codes: H04L 9/3213 using tickets or tokens, e.g. Kerberos network architectures or network communication protocols for entities authentication using tickets in a packet data network H04L63/0807
• CPC Class Codes: H04L 9/3247 involving digital signatures
• CPC Class Codes: H04W 12/04 : Key management, e.g. using generic bootstrapping architecture [GBA]
• CPC Class Codes: H04W 4/08 : User group management
• CPC Class Codes: Y02D 10/00 : Energy efficient computing, e.g. low power processors, power management or thermal management
• CPC Class Codes: Y04S 40/00 : Systems for electrical power generation, transmission, distribution or end-user application management characterised by the use of communication or information technologies, or communication or information technology specific aspects supporting them
• CPC Class Codes: Y04S 40/20 : Information technology specific aspects, e.g. CAD, simulation, modelling, system security

Abstract

Various approaches for deployment and use of configurable edge computing platforms are described. In an edge computing system, an edge computing device includes hardware resources that can be composed from a configuration of chiplets, as the chiplets are disaggregated for selective use and deployment (for compute, acceleration, memory, storage, or other resources). In an example, configuration operations are performed to: identify a condition for use of the hardware resource, based on an edge computing workload received at the edge computing device; obtain, determine, or identify properties of a configuration for the hardware resource that are available to be implemented with the chiplets, with the configuration enabling the hardware resource to satisfy the condition for use of the hardware resource; and compose the chiplets into the configuration, according to the properties of the configuration, to enable the use of the hardware resource for the edge computing workload.

Claims

1. An edge computing device in an edge computing system, comprising:

at least one hardware resource, the hardware resource composed from a configuration of a plurality of chiplets, wherein the chiplets provide disaggregated portions of the hardware resource; and

circuitry configured to perform operations to;

identify a condition for use of the hardware resource, based on an edge computing workload received at the edge computing device;

obtain properties of a configuration for the hardware resource that are available to be implemented with the chiplets, the configuration enabling the hardware resource to satisfy the condition for use of the hardware resource; and

compose the chiplets into the configuration, according to the properties of the configuration, to enable the use of the hardware resource for the edge computing workload.

2. The edge computing device of claim 1, the circuitry further configured to perform operations to:

identify the hardware resource as available in the edge computing device to be used for the edge computing workload according to the configuration.

3. The edge computing device of claim 2, the circuitry further configured to perform operations to:

verify the configuration of the chiplets in the hardware resource, wherein the chiplets provide attestation for the use of the hardware resource with the edge computing workload according to the configuration.

4. The edge computing device of claim 1, wherein the hardware resource comprises at least one of:

compute resources;

pooled acceleration resources;

pooled memory resources;

or

pooled storage resources.

5. The edge computing device of claim 4,

wherein composition of the chiplets into the configuration produces a virtual edge appliance for handling the edge computing workload on behalf of a tenant, and

wherein the virtual edge appliance provides use of:

a portion of accelerator components from the pooled acceleration resources, a portion of compute components from the compute resources, a portion of the memory from the pooled memory resources, and a portion of the storage from the pooled storage resources.

6. The edge computing device of claim 1,

wherein the hardware resource comprises a configurable resource, and

wherein the circuitry is further configured to identify a configuration for the configurable resource and load the configuration onto the configurable resource.

7. The edge computing device of claim 6, wherein the configurable resource comprises field-programmable gate array (FPGA) circuitry.

8. The edge computing device of claim 1, the circuitry configured to perform operations to:

determine and save the properties of the configuration for the hardware resource that are available to be implemented with the chiplets; and

reuse the configuration with a subsequent edge computing workload, in connection with the use of the hardware resource for the subsequent edge computing workload.

9. The edge computing device of claim 1, wherein the circuitry is provided from a modular fabric, wherein the modular fabric is configured to coordinate the chiplets for use as a virtual computing platform.

10. The edge computing device of claim 9, wherein the hardware resource includes a chiplet I/O interface to the modular fabric, the chiplet I/O interface configured to enable registration of the properties of the chiplets, manage hardware partitions of the chiplets, and connect the chiplets to the modular fabric.

11. The edge computing device of claim 9, wherein the modular fabric is further configured to implement logic to:

perform discovery and registration of the chiplets;

identify available compositions of the chiplets; and

enable a user software stack to discover functionality provided for the chiplets.

12. The edge computing device of claim 9, wherein the modular fabric is further configured to implement logic to:

create a virtual partition of at least a portion of the chiplets based on association with a particular tenant, and a service level agreement associated with the particular tenant;

wherein the circuitry is further configured to map a portion of the chiplets of the hardware resource to the virtual partition.

13. The edge computing device of claim 9, wherein the modular fabric is further configured to implement logic to:

verify attestation of the chiplets, using root of trust logic implemented by the modular fabric, based on attestation provided from the respective chiplets; and

verify an identity and configuration of the respective chiplets.

14. The edge computing device of claim 9, wherein the modular fabric is further configured to implement logic to:

monitor and enforce the configuration of the chiplets in the hardware resource, based on a composed node flavor of the virtual computing platform for a particular workload to be processed by the edge computing device.

15. The edge computing device of claim 14,

wherein the composed node flavor is one of a plurality of composed node flavors,

wherein respective node flavors of the plurality of composed node flavors correspond to respective types of workloads to be processed by the edge computing device, and

wherein the respective node flavors provide variations in type and amount of resources used for processing the respective types of workloads.

16. A method performed by an edge computing device, the edge computing device comprising at least one hardware resource composed from a configuration of a plurality of chiplets, and the method comprising:

identifying a condition for use of the hardware resource, based on an edge computing workload received at the edge computing device;

obtain properties of a configuration for the hardware resource that is available to be implemented with the chiplets, the configuration enabling the hardware resource to satisfy the condition for use of the hardware resource; and

composing the chiplets into the configuration, according to the properties of the configuration, to enable the use of the hardware resource for the edge computing workload.

17. The method of claim 16, further comprising:

identifying the hardware resource as available in the edge computing device to be used for the edge computing workload according to the configuration; and

verifying the configuration of the chiplets in the hardware resource, wherein the chiplets provide attestation for the use of the hardware resource with the edge computing workload according to the configuration.

18. The method of claim 16, wherein composition of the chiplets into the configuration produces a virtual edge appliance for handling the edge computing workload on behalf of a tenant, and

wherein the virtual edge appliance provides use of:

a portion of accelerator components from pooled acceleration resources, a portion of compute components from compute resources, a portion of memory from pooled memory resources, and a portion of storage from pooled storage resources.

19. The method of claim 16, wherein the hardware resource comprises a configurable resource, the method further comprising:

identifying a configuration for the configurable resource; and

loading the configuration onto the configurable resource.

20. The method of claim 16, wherein operations are performed by a modular fabric, wherein the modular fabric is configured to coordinate the chiplets for use as a virtual computing platform, and wherein the hardware resource includes a chiplet I/O interface to the modular fabric, the chiplet I/O interface configured to enable registration of the properties of the chiplets, manage hardware partitions of the chiplets, and connect the chiplets to the modular fabric.

21. The method of claim 20, wherein the modular fabric is further configured to implement logic to:

perform discovery and registration of the chiplets;

identify available compositions of the chiplets; and

enable a user software stack to discover functionality provided for the chiplets.

22. The method of claim 20,

wherein the modular fabric is further configured to implement logic to create a virtual partition of at least a portion of the chiplets based on association with a particular tenant, and a service level agreement associated with the particular tenant, and

wherein the method further comprises mapping a portion of the chiplets of the hardware resource to the virtual partition.

23. The method of claim 20, wherein the modular fabric is further configured to implement logic for at least one of:

verify attestation of the chiplets, using root of trust logic implemented by the modular fabric, based on attestation provided from the chiplets;

verify an identity and configuration of the respective chiplets;

or

monitor and enforce the configuration of the chiplets in the hardware resource, based on a composed node flavor of the virtual computing platform for a particular workload to be processed by the edge computing device.

24. At least one non-transitory machine-readable medium comprising instructions, wherein the instructions, when implemented by a processing circuitry of a computing device, cause the processing circuitry to perform operations that:

identify a condition for use of a hardware resource composed from a configuration of a plurality of chiplets, based on an edge computing workload;

obtain properties of a configuration for the hardware resource that is available to be implemented with the chiplets, the configuration enabling the hardware resource to satisfy the condition for use of the hardware resource; and

compose the chiplets into the configuration, according to the properties of the configuration, to enable the use of the hardware resource for the edge computing workload.

25. The machine-readable medium of claim 24, the processing circuitry further configured to perform operations to:

identify the hardware resource as available to be used for the edge computing workload according to the configuration; and

verify the configuration of the chiplets in the hardware resource, wherein the chiplets provide attestation for the use of the hardware resource with the edge computing workload according to the configuration;

wherein the chiplets provide disaggregated portions of the hardware resource.