Multi-protocol communication fabric control

Patent metadata

Field	Value
Publication number	US-11973650-B2
Application number	US-202016858005-A
Country	US
Kind code	B2
Filing date	Apr 24, 2020
Priority date	Apr 25, 2019
Publication date	Apr 30, 2024
Grant date	Apr 30, 2024

Abstract

Deployment of arrangements of computing components coupled over a communication fabric are presented herein. In one example, a method includes detecting first computing components communicatively coupled to a first communication fabric having a first communication fabric type, and detecting second computing components communicatively coupled to a second communication fabric having a second communication fabric type. The method also includes receiving user commands to form compute units among a pool of computing components comprising the first computing components and the second computing components. Based at least on the user commands, the method includes forming the compute units for use by one or more users.

First claim

What is claimed is:

1. A method comprising:
detecting first computing components communicatively coupled to a first communication fabric that conforms to a first communication fabric protocol;
detecting second computing components communicatively coupled to a second communication fabric that conforms to a second communication fabric protocol different than the first communication fabric protocol;
receiving user commands to form a compute unit among a pool of computing components comprising the first computing components and the second computing components, wherein the user commands indicate to form the compute unit as having selected computing components from both the first communication fabric and the second communication fabric;
based at least on the user commands, forming the compute unit by at least forming a logical domain that spans the first communication fabric and the second communication fabric and includes the selected computing components;
initiating a conversion unit to interwork communications for the compute unit among first selected computing components that use the first communication fabric protocol and second selected computing components that use the second communication fabric protocol; and
based at least on a policy affecting the compute unit, migrating a workload for the compute unit by at least adding, into the compute unit, additional computing components that use the second communication fabric protocol and removing, from the compute unit, the first selected computing components that use the first communication fabric protocol.

2. The method of claim 1 , further comprising:
presenting the first computing components and the second computing components in a control interface as a shared pool of computing components available to compose into arbitrary compute units having computing components selected from both the first communication fabric and the second communication fabric; and
receiving user commands via the control interface.

3. The method of claim 1 , further comprising:
forming a first logical partition in the first communication fabric and a second logical partition in the second communication fabric to include the selected computing components of the compute unit into the logical domain.

4. The method of claim 1 , wherein the conversion unit is configured to interwork different ones among communication fabric types, communication protocol types, or communication protocol versions among the selected computing components of the compute unit.

5. The method of claim 1 , further comprising:
based at least on a first configuration inhibiting spanning across different communication fabric types, indicating to a user that at least one of the computing components of the pool of computing components cannot be included among the selected computing components;
based at least on a second configuration allowing spanning across different communication fabric types, indicating to the user that at least another of the computing components of the pool of computing components can be included among the selected computing components.

6. The method of claim 1 , wherein the first communication fabric comprises a communication fabric conforming to a Peripheral Component Interconnect Express (PCIe) communication protocol standard, and wherein the second communication fabric comprises a communication fabric conforming to at least one among a Gen-Z, InfiniBand, NVMe, Ethernet, Serial Attached SCSI (SAS), FibreChannel, Thunderbolt, Serial Attached ATA Express (SATA Express), Compute Express Link (CXL), Cache Coherent Interconnect for Accelerators (CCIX), Open Coherent Accelerator Processor Interface (OpenCAPI), Wi-Fi, cellular wireless communication fabric, and different version of PCIe fabric.

7. The method of claim 1 , wherein a management processor is communicatively coupled to the first communication fabric, and wherein the management processor is configured to communicate over the second communication fabric via the first communication fabric.

8. The method of claim 1 , further comprising:
performing a discovery process to identify the first computing components as associated with the first communication fabric protocol, and identify the second computing components as associated with the second communication fabric protocol; and
maintaining one or more data structures which indicates a communication fabric protocol for each of the first computing components and each of the second computing components.

9. The method of claim 1 , wherein the first communication fabric protocol and the second communication fabric protocol each comprise a different protocol type or different protocol version conforming to at least one among a Peripheral Component Interconnect Express (PCIe), Gen-Z, InfiniBand, NVMe, Ethernet, Serial Attached SCSI (SAS), FibreChannel, Thunderbolt, Serial Attached ATA Express (SATA Express), Compute Express Link (CXL), Cache Coherent Interconnect for Accelerators (CCIX), Open Coherent Accelerator Processor Interface (OpenCAPI) Wi-Fi, and cellular wireless communication fabric.

10. An apparatus, comprising:
one or more computer readable storage media; and
program instructions stored on the one or more computer readable storage media that, based on being read and executed by a processing system, direct the processing system to at least:
detect first computing components communicatively coupled to a first communication fabric that conforms to a first communication fabric protocol;
detect second computing components communicatively coupled to a second communication fabric that conforms to a second communication fabric protocol different than the first communication fabric protocol;
receive user commands to form a compute unit among a pool of computing components comprising the first computing components and the second computing components, wherein the user commands indicate to form the compute unit as having selected computing components from both the first communication fabric and the second communication fabric;
based at least on the user commands, form the compute unit by at least forming a logical domain that spans the first communication fabric and the second communication fabric and includes the selected computing components;
initiate a conversion unit to interwork communications for the compute unit among first selected computing components that use the first communication fabric protocol and second selected computing components that use the second communication fabric protocol; and
based at least on a policy affecting the compute unit, migrate a workload for the compute unit by at least adding, into the compute unit, additional computing components that use the second communication fabric protocol and removing, from the compute unit, the first selected computing components that use the first communication fabric protocol.

11. The apparatus of claim 10 , wherein the conversion unit is configured to interwork different ones among communication fabric types, communication protocol types, or communication protocol versions among the selected computing components of the compute unit.

12. The apparatus of claim 10 , comprising further program instructions, based on being executed by the processing system, direct the processing system to at least:
based at least on a first configuration inhibiting spanning across different communication fabric types, indicate to a user that at least one of the computing components of the pool of computing components cannot be included among the selected computing components;
based at least on a second configuration allowing spanning across different communication fabric types, indicate to the user that at le

Assignees

• Liqid Inc

Inventors

• Cannata James Scott
• Clark Phillip
• Nagel Bryan
• Harris Henry Lee
• Schramm Bryan
• Puri Sumit

Classifications

• H04L69/18

  Multiprotocol handlers, e.g. single devices capable of handling multiple protocols · CPC title

• H04L69/08

  Protocols for interworking; Protocol conversion · CPC title

• H04L47/10

  Flow control; Congestion control · CPC title

• H04L67/1097

  for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS] · CPC title

• H04L43/20

  the monitoring system or the monitored elements being virtualised, abstracted or software-defined entities, e.g. SDN or NFV · CPC title