US 11,853,783 B1
Identifying hosts for dynamically enabling specified features when resuming operation of a virtual compute instance
Mohamed A Aboubakr, Issaquah, WA (US); Deepak Bhardwaj, Seattle, WA (US); Elton Victor Pinto, Seattle, WA (US); Ziran Nie, Seattle, WA (US); Akshay Bharat Shah, Redmond, WA (US); Ayush Jitendra Sanghvi, Seattle, WA (US); Michael A Ibarra, Everett, WA (US); Venkat Srinath Vutphala, Bothell, WA (US); and Shivanischal Anand, Bellevue, WA (US)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on Mar. 29, 2021, as Appl. No. 17/216,482.
Int. Cl. G06F 9/455 (2018.01)
CPC G06F 9/45545 (2013.01) [G06F 9/45558 (2013.01); G06F 2009/4557 (2013.01); G06F 2009/45562 (2013.01); G06F 2009/45575 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system, comprising:
a control plane for a virtual compute service, wherein the control plane comprises at least one processor and a memory, and wherein the control plane is configured to:
receive a request to start operation of a virtual compute instance that was previously running on a first host system of the virtual compute service;
identify one or more features that were previously enabled at runtime as specified in a request received via an interface of the virtual computing service, wherein enablement of the one or more features provided a customized execution environment for the virtual compute instance at the first host system, and wherein the one or more features were enabled via the interface prior to stopping operation of the virtual compute instance;
select, from a plurality of available host systems for the virtual compute instance, a second host system that supports at least the one or more features;
cause the one or more features that were previously enabled at runtime at the first host system and are currently disabled at the second host system to be enabled at the second host system, providing the same customized execution environment at the second host system for the virtual compute instance; and
cause operation of the virtual compute instance to start at the second host system according to an in-memory state stored responsive to stopping operation of the virtual compute instance.