Dynamic grouping and repurposing of general purpose links in disaggregated datacenters
Embodiments are provided herein for efficient component communication and resource utilization in a disaggregated computing system. An application programming interface (API) is used to submit a communications request to a communications manager, the communications request associated with a connection between a first plurality of computing elements of a first type and a second plurality of computing elements of a second type. The connection between ones of the first plurality of computing elements and the second plurality of computing elements is dynamically rewired according to a grouping scheme computed using information contained within the communications request.
1. A method for efficient component communication and resource utilization in a disaggregated computing system, by a processor, comprising:
using an application programming interface (API) to submit a communications request to a communications manager, the communications request associated with a connection between a first plurality of computing elements of a first type and a second plurality of computing elements of a second type; and
dynamically rewiring the connection between ones of the first plurality of computing elements and the second plurality of computing elements according to a grouping scheme computed using information contained within the communications request.
2. The method of claim 1 , wherein the first plurality of computing elements comprise processors and the second plurality of computing elements comprise memory elements.
3. The method of claim 1 , wherein the communications request is submitted upon detecting a triggering event has occurred.
4. The method of claim 3 , wherein the triggering event comprises identifying that an observed pressure threshold associated with at least one of a bandwidth, a latency, and a throughput has been reached.
5. The method of claim 3 , wherein the triggering event comprises identifying an observed workload traffic pattern, the observed workload pattern historically learned from specific workloads.
6. The method of claim 3 , wherein the triggering event comprises a user directive.
7. The method of claim 1 , further including, pursuant to computing the grouping scheme, identifying respective locations of the ones of the first and second pluralities of computing elements; and
determining a most efficient grouping of the ones of the first and second pluralities of computing elements based on the identified respective locations for forming the rewired connection according to the information contained within the communications request comprising respective data locations and data travel distances between the first and second pluralities of computing elements.
8. A system for efficient component communication and resource utilization in a disaggregated computing system, comprising:
a first plurality of computing elements of a first type; and
a second plurality of computing elements of a second type; wherein at least one of the first or second plurality of computing elements comprises a processor; wherein the processor:
uses an application programming interface (API) to submit a communications request to a communications manager, the communications request associated with a connection between the first plurality of computing elements of the first type and the second plurality of computing elements of the second type; and
dynamically rewires the connection between ones of the first plurality of computing elements and the second plurality of computing elements according to a grouping scheme computed using information contained within the communications request.
9. The system of claim 8 , wherein the first plurality of computing elements comprise processors and the second plurality of computing elements comprise memory elements.
10. The system of claim 8 , wherein the communications request is submitted upon detecting a triggering event has occurred.
11. The system of claim 10 , wherein the triggering event comprises identifying that an observed pressure threshold associated with at least one of a bandwidth, a latency, and a throughput has been reached.
12. The system of claim 10 , wherein the triggering event comprises identifying an observed workload traffic pattern, the observed workload pattern historically learned from specific workloads.
13. The system of claim 10 , wherein the triggering event comprises a user directive.
14. The system of claim 8 , wherein the processor, pursuant to computing the grouping scheme, identifies respective locations of the ones of the first and second pluralities of computing elements; and
determines a most efficient grouping of the ones of the first and second pluralities of computing elements based on the identified respective locations for forming the rewired connection according to the information contained within the communications request comprising respective data locations and data travel distances between the first and second pluralities of computing elements.
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 an application programming interface (API) to submit a communications request to a communications manager, the communications request associated with a connection between a first plurality of computing elements of a first type and a second plurality of computing elements of a second type; and
an executable portion that dynamically rewires the connection between ones of the first plurality of computing elements and the second plurality of computing elements according to a grouping scheme computed using information contained within the communications request.
16. The computer program product of claim 15 , wherein the first plurality of computing elements comprise processors and the second plurality of computing elements comprise memory elements.
17. The computer program product of claim 15 , wherein the communications request is submitted upon detecting a triggering event has occurred.
18. The computer program product of claim 17 , wherein the triggering event comprises identifying that an observed pressure threshold associated with at least one of a bandwidth, a latency, and a throughput has been reached.
19. The computer program product of claim 17 , wherein the triggering event comprises identifying an observed workload traffic pattern, the observed workload pattern historically learned from specific workloads.
20. The computer program product of claim 17 , wherein the triggering event comprises a user directive.
21. The computer program product of claim 15 , further including an executable portion that, pursuant to computing the grouping scheme, identifies respective locations of the ones of the first and second pluralities of computing elements; and
an executable portion that determines a most efficient grouping of the ones of the first and second pluralities of computing elements based on the identified respective locations for forming the rewired connection according to the information contained within the communications request comprising respective data locations and data travel distances between the first and second pluralities of computing elements.