US 12,074,806 B2
Nondominant resource management for edge multi-tenant applications
S M Iftekharul Alam, Hillsboro, OR (US); Satish Chandra Jha, Portland, OR (US); Ned M. Smith, Beaverton, OR (US); Vesh Raj Sharma Banjade, Portland, OR (US); Kshitij Arun Doshi, Tempe, AZ (US); Francesc Guim Bernat, Barcelona (ES); Arvind Merwaday, Beaverton, OR (US); Kuilin Clark Chen, Portland, OR (US); and Christian Maciocco, Portland, OR (US)
Assigned to Intel Corporation, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on Apr. 1, 2022, as Appl. No. 17/711,921.
Prior Publication US 2022/0231964 A1, Jul. 21, 2022
Int. Cl. H04L 47/72 (2022.01); H04L 47/74 (2022.01)
CPC H04L 47/72 (2013.01) [H04L 47/74 (2013.01)] 22 Claims
OG exemplary drawing
 
1. A system for managing dominant edge/cloud computing resources, the system comprising:
memory controller processing circuitry; and
a memory that includes instructions, the instructions, when executed by the memory controller processing circuitry, cause the memory controller processing circuitry to:
receive a computing job request from a client computing device at a resource manager, the computing job request including a dominant resource request for a dominant resource and a non-dominant resource request for a non-dominant resource, the dominant resource including a first computing resource whose specifications constrain an execution of the computing job request;
determine a proportional allocation of the non-dominant resource request and the dominant resource request;
allocate, based on the proportional allocation, the non-dominant resource request to the non-dominant resource and the dominant resource request to the dominant resource;
issue a request to cause the execution of the computing job request at the non-dominant resource and the dominant resource;
receive a resource telemetry generated during execution of the computing job request, the resource telemetry including a non-dominant resource usage telemetry generated at the non-dominant resource and a dominant resource usage telemetry generated at the dominant resource during execution of the computing job request;
determine, based on the resource telemetry, that the computing job request includes a non-dominant resource usage attack;
issue a request to cause the computing job request to be halted based on the resource telemetry;
add the client computing device to a client device blocklist; and
prevent a subsequent client job request from the client computing device.