Remote latency calls for latency clustered compute instances
A device receives, from a first compute instance, a Remote Latency Call, that includes a target latency measure, a function, and one or more arguments for the function. The device clusters a plurality of compute instances into multiple clusters of compute instances based on a respective latency measure associated with each compute instance of the plurality of compute instances, and determines a first cluster of compute instances from the multiple clusters of compute instances that satisfies the target latency measure. The device causes a second compute instance within the first cluster of compute instances to execute the function, using the one or more arguments, and return one or more results.
1. A device, comprising:
at least one communication interface configured to receive, from a first compute instance, a Remote Latency Call, that includes a target latency measure, a function, and one or more arguments for the function; and
a processor configured to:
cluster a plurality of compute instances into multiple clusters of compute instances based on a respective latency measure associated with each compute instance of the plurality of compute instances,
determine a first cluster of compute instances from the multiple clusters of compute instances that satisfies the target latency measure, and
cause a second compute instance within the first cluster of compute instances to execute the function, using the one or more arguments, and return one or more results.
2. The device of claim 1 , wherein, when causing the second compute instance to execute the function, the processor is further configured to:
migrate, via the at least one communication interface, code associated with the function to the second compute instance within the first cluster of compute instances, and
pass the one or more arguments to the second compute instance for execution of the function.
3. The device of claim 1 , wherein the processor is further configured to:
receive the one or more results of execution of the function from the second compute instance, and
pass the one or more results to the first compute instance.
4. The device of claim 1 , wherein the processor is further configured to:
select the second compute instance from the first cluster of compute instances by applying a round robin selection policy.
5. The device of claim 1 , wherein the processor is further configured to:
select the second compute instance from the first cluster of compute instances based on a current processing load associated with each of the compute instances within the first cluster of compute instances.
6. The device of claim 1 , wherein the target latency measure comprises a target round trip time (RTT).
7. The device of claim 6 , wherein, when clustering the plurality of compute instances and determining the first cluster of compute instances, the processor is further configured to:
assign multiple RTT classes to the multiple clusters of compute instances,
identify a first RTT class, of the multiple RTT classes, based on the target RTT, and
map the identified first RTT class to the first cluster of compute instances.
8. The device of claim 6 , wherein the respective latency measure associated with each compute instance comprises an RTT.
9. A method, comprising:
receiving, from a first compute instance, a Remote Latency Call, that includes a target latency measure, a function, and one or more arguments for the function;
clustering a plurality of compute instances into multiple clusters of compute instances based on a respective latency measure associated with each compute instance of the plurality of compute instances;
determining a first cluster of compute instances from the multiple clusters of compute instances that satisfies the target latency measure; and
causing a second compute instance within the first cluster of compute instances to execute the function, using the one or more arguments, and return one or more results.
10. The method of claim 9 , wherein causing the second compute instance to execute the function further comprises:
migrating code associated with the function to the second compute instance within the first cluster of compute instances; and
passing the one or more arguments to the second compute instance for execution of the function.
11. The method of claim 9 , further comprising:
receiving the one or more results of execution of the function from the second compute instance; and
passing the one or more results to the first compute instance.
12. The method of claim 9 , further comprising:
selecting the second compute instance from the first cluster of compute instances by applying a round robin selection policy.
13. The method of claim 9 , further comprising:
selecting the second compute instance from the first cluster of compute instances based on a current processing load associated with each of the compute instances within the first cluster of compute instances.
14. The method of claim 9 , wherein the target latency measure comprises a target round trip time (RTT).
15. The method of claim 14 , wherein clustering the plurality of compute instances and determining the first cluster of compute instances further comprises:
assigning multiple RTT classes to the multiple clusters of compute instances;
identifying a first RTT class, of the multiple RTT classes, based on the target RTT; and
mapping the identified first RTT class to the first cluster of compute instances.
16. The method of claim 14 , wherein the respective latency measure associated with each compute instance comprises a RTT.
17. A non-transitory storage medium storing instructions executable by a device with one or more processors, wherein execution of the instructions causes the device to:
receive, from a first compute instance, a Remote Latency Call, that includes a target latency measure, a function, and one or more arguments for the function;
cluster a plurality of compute instances into multiple clusters of compute instances based on a respective latency measure associated with each compute instance of the plurality of compute instances;
determine a first cluster of compute instances from the multiple clusters of compute instances that satisfies the target latency measure; and
cause a second compute instance within the first cluster of compute instances to execute the function, using the one or more arguments, and return one or more results.
18. The non-transitory storage medium of claim 17 , wherein, when causing the selected second compute instance to execute the function, execution of the instructions further causes the device to:
migrate code associated with the function to the selected second compute instance within the first cluster of compute instances; and
pass the one or more arguments to the second compute instance for execution of the function.
19. The non-transitory storage medium of claim 17 , wherein the target latency measure comprises a target round trip time (RTT).
20. The non-transitory storage medium of claim 19 , wherein, when clustering the plurality of compute instances and determining the first cluster of compute instances, execution of the instructions further causes the device to:
assign multiple RTT classes to the multiple clusters of compute instances;
identify a first RTT class, of the multiple RTT classes, based on the target RTT; and
map the identified first RTT class to the first cluster of compute instances.