Performance-based hardware emulation in an on-demand network code execution system

A computer-implemented method comprising:

• receiving user-submitted code executable on an on-demand code execution system;
• determining, based at least in part on the user-submitted code, computing resource functionality associated with executing the user-submitted code on the on-demand code execution system;
• and
• in response to a request to execute the user-submitted code:
• obtaining first performance information regarding a first computing resource that physically implements the computing resource functionality;
• obtaining second performance information regarding a second computing resource that does not physically implement the computing resource functionality but that virtually emulates the computing resource functionality;
• determining, based at least in part on the first performance information and the second performance information, a recommended computing resource for executing the user-submitted code, the recommended computing resource being one of the first computing resources or the second computing resource;
• and
• providing a recommendation that includes the recommended computing resource to the on-demand code execution system, wherein the on-demand code execution system selects a computing resource for executing the user-submitted code based at least in part on the recommendation.

Non-transitory computer-readable media including computer-executable instructions that, when executed by an on-demand code execution system, cause the on-demand code execution system to:

• obtain user-submitted code executable on the on-demand code execution system;
• determine, based at least in part on the user-submitted code, computing resource functionality associated with executing the user-submitted code on the on-demand code execution system, wherein the on-demand code execution system does not receive a request to provide the computing resource functionality;
• identifying a plurality of computing resources that implement the computing resource functionality;
• in response to a request to execute the user-submitted code:
• identifying an available subset of the plurality of computing resources that either physically implement or virtually emulate the computing resource functionality;
• selecting, from the available subset, a recommended computing resource for executing the user-submitted code based at least in part on performance of individual computing resources in the available subset at providing the computing resource functionality;
• and
• executing the user-submitted code on the recommended computing resource.