US 11,695,840 B2
Dynamically routing code for executing
George Steven McPherson, Seattle, WA (US); Mehul A. Shah, Saratoga, CA (US); Supratik Chakraborty, San Jose, CA (US); Prajakta Datta Damle, San Jose, CA (US); Gopinath Duddi, San Jose, CA (US); and Anurag Windlass Gupta, Atherton, CA (US)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on Mar. 14, 2022, as Appl. No. 17/694,585.
Application 17/694,585 is a continuation of application No. 15/385,784, filed on Dec. 20, 2016, granted, now 11,277,494.
Claims priority of provisional application 62/426,584, filed on Nov. 27, 2016.
Prior Publication US 2022/0279049 A1, Sep. 1, 2022
Int. Cl. H04L 67/00 (2022.01); H04L 67/10 (2022.01); G06F 9/50 (2006.01); G06F 9/4401 (2018.01); H04L 67/01 (2022.01)
CPC H04L 67/34 (2013.01) [G06F 9/4411 (2013.01); G06F 9/5005 (2013.01); G06F 9/5072 (2013.01); H04L 67/01 (2022.05); H04L 67/10 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system, comprising:
at least one processor; and
a memory, storing program instructions that when executed by the at least one processor, cause the at least one processor to:
receive, at a network endpoint, source code of a job to transfer data from a data source to a on behalf of a client;
in response to the receipt of the source code:
evaluate the job to determine a criteria for the request to execute the source code, wherein the evaluating comprises:
identify use of one or more Graphics Processing Unit (GPUs) that are utilized via one or more application programming interfaces (APIs) as part of executing the source code and identifying that a driver is used to interact with the one or more GPUs in order to execute the source code;
evaluate respective capabilities of available computing resources with respect to the determined criteria to identify one or more computing resources from the available computing resources with the respective capabilities to execute the source code that satisfy the determined criteria including one or more GPUs;
procure the identified one or more computing resources with the respective capabilities to execute the source code, wherein the procuring causes configuration of a software image for a software container that implements operating system virtualization at the identified one or more computing resources to use the driver to interact with the one or more GPUs via the one or more APIs to execute the source code; and
initiate execution of the source code at the procured computing resources.