Dynamic component communication using general purpose links between respectively pooled together of like typed devices in disaggregated datacenters

• US 10,671,557 B2
• Filed: 09/25/2018
• Issued: 06/02/2020
• Est. Priority Date: 09/25/2018
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Li, Min, Bivens, John A., Mahindru, Ruchi, Salapura, Valentina, Schenfeld, Eugen
• Primary Examiner(s): Peyton, Tammara R
• Application Number:
  US16/141,878
• Publication Number:
  US 20200097428A1
• Time in Patent Office: 616 Days
• CPC Class Codes: G06F 12/109 for multiple virtual address spaces, e.g. segmentation G06F12/1036 takes precedence
• CPC Class Codes: G06F 13/4022 using switching circuits, e.g. switching matrix, connection or expansion network G06F13/4009 takes precedence
• CPC Class Codes: G06F 2212/152 Virtualized environment, e.g. logically partitioned system
• CPC Class Codes: G06F 2212/154 Networked environment
• CPC Class Codes: H04B 10/27 Arrangements for networking
• CPC Class Codes: H04Q 11/0005 Switch and router aspects
• CPC Class Codes: H04Q 11/0062 Network aspects
• CPC Class Codes: H04Q 2011/0086 Network resource allocation, dimensioning or optimisation
• CPC Class Codes: H04Q 2213/1301 Optical transmission, optical switches

Abstract

Embodiments are provided herein for efficient component communication and resource utilization in a disaggregated computing system. A general purpose link is provided between a plurality of devices in the disaggregated computing system such that the general purpose link is used to connect the plurality of devices. For those of the plurality of devices communicating with one another within a same pool, the connection of the general purpose link is established using a backplane to facilitate the communication; and for those of the plurality of devices communicating with one another within differing pools, the connection of the general purpose link is established through an optical switching device to facilitate the communication.

Claims

1. A method for efficient component communication and resource utilization in a disaggregated computing system, by a processor, comprising:

providing a general purpose link between a plurality of devices respectively pooled together in pools of like-typed devices, the plurality of devices comprising individual physical hardware components used to dynamically construct the disaggregated computing system, wherein the general purpose link is used to connect the plurality of devices;

wherein;

for those of the plurality of devices communicating with one another within a same pool, the connection of the general purpose link is established using a backplane to facilitate the communication; and

for those of the plurality of devices communicating with one another within differing pools, the connection of the general purpose link is established through an optical switching device to facilitate the communication.

2. The method of claim 1, wherein the plurality of devices are selected from a list comprising processors, memory devices, storage devices, and accelerators.

3. The method of claim 1, wherein physical pins of a first one of the plurality of devices are used to connect to the general purpose link and communicate with respective other ones of the plurality of devices such that the physical pins may be reused to communicate with differing types of devices.

4. The method of claim 1, wherein respective ones of the plurality of devices within the same pool share a bandwidth of the connection of the general purpose link when communicating with respective ones of the plurality of devices in another pool.

5. The method of claim 1, further including rerouting data transferred between ones of the plurality of devices through an alternate general purpose link using the optical switching device according to at least one of a plurality of factors.

6. The method of claim 5, wherein the plurality of factors include at least a link bandwidth utilization, a link throughput, and a rewiring cost.

7. The method of claim 1, further including reallocating ones of the plurality of devices from the differing pools to the same pool upon determining the general purpose link is already established to respective other ones of the plurality of devices within the same pool.

8. A system for efficient component communication and resource utilization in a disaggregated computing system, comprising:

a plurality of devices respectively pooled together in pools of like-typed devices, the plurality of devices comprising individual physical hardware components used to dynamically construct the disaggregated computing system;

wherein at least one of the plurality of devices comprises a processor executing instructions stored in a memory device;

wherein the processor;

uses a general purpose link between the plurality of devices to connect the plurality of devices;

wherein;

for those of the plurality of devices communicating with one another within a same pool, the connection of the general purpose link is established using a backplane to facilitate the communication; and

for those of the plurality of devices communicating with one another within differing pools, the connection of the general purpose link is established through an optical switching device to facilitate the communication.

9. The system of claim 8, wherein the plurality of devices are selected from a list comprising processors, memory devices, storage devices, and accelerators.

10. The system of claim 8, wherein physical pins of a first one of the plurality of devices are used to connect to the general purpose link and communicate with respective other ones of the plurality of devices such that the physical pins may be reused to communicate with differing types of devices.

11. The system of claim 8, wherein respective ones of the plurality of devices within the same pool share a bandwidth of the connection of the general purpose link when communicating with respective ones of the plurality of devices in another pool.

12. The system of claim 8, wherein the processor reroutes data transferred between ones of the plurality of devices through an alternate general purpose link using the optical switching device according to at least one of a plurality of factors.

13. The system of claim 12, wherein the plurality of factors include at least a link bandwidth utilization, a link throughput, and a rewiring cost.

14. The system of claim 8, wherein the processor reallocates ones of the plurality of devices from the differing pools to the same pool upon determining the general purpose link is already established to respective other ones of the plurality of devices within the same pool.

15. A computer program product for efficient component communication and resource utilization in a disaggregated computing system, by a processor, the computer program product embodied on a non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:

an executable portion that uses a general purpose link between a plurality of devices respectively pooled together in pools of like-typed devices, the plurality of devices comprising individual physical hardware components used to dynamically construct the disaggregated computing system, wherein the general purpose link is used to connect the plurality of devices, wherein the general purpose link is used to connect the plurality of devices;

wherein;

for those of the plurality of devices communicating with one another within a same pool, the connection of the general purpose link is established using a backplane to facilitate the communication; and

for those of the plurality of devices communicating with one another within differing pools, the connection of the general purpose link is established through an optical switching device to facilitate the communication.

16. The computer program product of claim 15, wherein the plurality of devices are selected from a list comprising processors, memory devices, storage devices, and accelerators.

17. The computer program product of claim 15, wherein physical pins of a first one of the plurality of devices are used to connect to the general purpose link and communicate with respective other ones of the plurality of devices such that the physical pins may be reused to communicate with differing types of devices.

18. The computer program product of claim 15, wherein respective ones of the plurality of devices within the same pool share a bandwidth of the connection of the general purpose link when communicating with respective ones of the plurality of devices in another pool.

19. The computer program product of claim 15, further including an executable portion that reroutes data transferred between ones of the plurality of devices through an alternate general purpose link using the optical switching device according to at least one of a plurality of factors.

20. The computer program product of claim 19, wherein the plurality of factors include at least a link bandwidth utilization, a link throughput, and a rewiring cost.

21. The computer program product of claim 15, further including an executable portion that reallocates ones of the plurality of devices from the differing pools to the same pool upon determining the general purpose link is already established to respective other ones of the plurality of devices within the same pool.