US 12,132,668 B2
Network-aware resource allocation
Ali Tariq, Palo Alto, CA (US); Lianjie Cao, Milpitas, CA (US); Faraz Ahmed, Milpitas, CA (US); and Puneet Sharma, Palo Alto, CA (US)
Assigned to Hewlett Packard Enterprise Development LP, Spring, TX (US)
Filed by Hewlett Packard Enterprise Development LP, Spring, TX (US)
Filed on May 3, 2023, as Appl. No. 18/311,430.
Application 18/311,430 is a continuation of application No. 17/468,517, filed on Sep. 7, 2021, granted, now 11,665,106.
Prior Publication US 2023/0275848 A1, Aug. 31, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. H04L 47/78 (2022.01); H04L 43/0882 (2022.01); H04L 43/16 (2022.01); H04L 47/70 (2022.01); H04L 47/762 (2022.01); H04L 47/80 (2022.01)
CPC H04L 47/803 (2013.01) [H04L 43/0882 (2013.01); H04L 43/16 (2013.01); H04L 47/762 (2013.01); H04L 47/781 (2013.01); H04L 47/822 (2013.01)] 26 Claims
OG exemplary drawing
 
1. A computing device comprising:
a memory; and
one or more processors that are configured to execute machine readable instructions stored in the memory that operate the one or more processors to:
determine a resource efficiency value of a plurality of resource containers in a distributed network;
compare the resource efficiency value to a threshold resource efficiency value;
when the resource efficiency value is greater than or equal to the threshold resource efficiency value, generate a distributed resource configuration that includes a resource upscaling process;
when the resource efficiency value is less than the threshold resource efficiency value, generate the distributed resource configuration that includes a resource outscaling process;
transmit the distributed resource configuration to a resource allocation platform of the distributed network, wherein the resource allocation platform updates allocation of the plurality of resource containers in the distributed network in accordance with the distributed resource configuration;
convert the distributed resource configuration to a weighted graph, wherein the weighted graph comprises a first number of parameter server nodes and a first number of worker nodes; and
using the weighted graph, determine a placement of a compute node or a resource adjustment at node-level or component-level.