US 12,346,734 B2
Metrics and security-based accelerator service rescheduling and auto-scaling using a programmable network device
Mikko Ylinen, Lempaala (FI); Ismo Puustinen, Helsinki (FI); Reshma Lal, Portland, OR (US); and Soham Jayesh Desai, Rochester, MN (US)
Assigned to INTEL CORPORATION, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on Sep. 22, 2021, as Appl. No. 17/482,155.
Prior Publication US 2022/0012095 A1, Jan. 13, 2022
Int. Cl. G06F 9/50 (2006.01); G06F 9/48 (2006.01); G06F 21/53 (2013.01); H04L 9/08 (2006.01); H04L 9/40 (2022.01)
CPC G06F 9/505 (2013.01) [G06F 9/4843 (2013.01); G06F 21/53 (2013.01); H04L 9/085 (2013.01); H04L 63/0428 (2013.01); H04L 63/08 (2013.01); G06F 2209/505 (2013.01); G06F 2209/509 (2013.01); G06F 2221/2149 (2013.01)] 19 Claims
OG exemplary drawing
 
1. An apparatus comprising:
one or more processor hardware devices comprising:
a control plane scheduler of a service mesh that is to manage microservices of a service and is to schedule workloads of the service on one or more hardware resources in a datacenter comprising the one or more processor hardware devices; and
a scheduler extender communicably coupled to the control plane scheduler, the scheduler extender executed inside of a trusted execution environment (TEE) to isolate the scheduler extender, and wherein the scheduler extender further comprises:
a metric analyzer to:
analyze telemetry data corresponding to communication links between the microservices of the service managed by the service mesh;
determine, based on analysis of the telemetry data, that a workload of the service can be accelerated by offload to a hardware accelerator device of the one or more hardware resources in the datacenter; and
identify, based on the analysis of the telemetry data, a type of the hardware accelerator device to receive the offload of the workload of the service; and
a resource allocator communicably coupled to the metric analyzer, the resource allocator to:
generate a scaling request to cause the hardware accelerator device corresponding to the type to be allocated to a cluster of hardware devices of the one or more hardware resources configured for the service;
transmit the scaling request to a programmable network device managing the hardware accelerator device, the programmable network device to allocate the hardware accelerator device to the cluster and to register the hardware accelerator device with the service mesh; and
schedule the workload of the service to the hardware accelerator device.