Managing computer workloads across distributed computing clusters

1. A system comprising:

a processor; and

a memory including instructions for a workload assignment module, the instructions being executable by the processor for causing the processor to:

receive a request from a workload manager for identifying a computing cluster to which to assign a workload;

receive historical information from a particular computing cluster, the historical information including a set of statistical distributions associated with a set of software components executed by the particular computing cluster during a prior timespan, wherein each statistical distribution in the set of statistical distributions indicates how many times the particular computing cluster executed different numbers of replicas of an associated software component during the prior timespan;

determine that the workload is to be assigned to the particular computing cluster among a plurality of computing clusters based on a metric derived from the set of statistical distributions in the historical information; and

transmit a response to the workload manager for causing the workload manager to assign the workload to the particular computing cluster.

2. The system of claim 1 , wherein the historical information includes statistical information about replicas of a software component deployed by the particular computing cluster during the prior timespan.

3. The system of claim 2 , wherein the software component is a serverless function or a microservice.

4. The system of claim 1 , wherein the plurality of computing clusters are edge clusters at physical edges of a distributed computing environment.

5. The system of claim 1 , wherein the workload manager is part of a telecommunication infrastructure that includes the plurality of computing clusters, and wherein the workload assignment module is external to the telecommunication infrastructure and in communication with the workload manager via the Internet.

6. The system of claim 1 , wherein the instructions are further executable by the processor for causing the processor to:

estimate resource usage by the particular computing cluster based on the workload and the metric; and

select the particular computing cluster from among the plurality of computing clusters in response to determining that the estimated resource usage is less than a resource-usage threshold for the particular computing cluster.

7. The system of claim 6 , wherein the instructions are further executable by the processor for causing the processor to estimate the resource usage associated with the particular computing cluster by:

determining a first mean of a first statistical distribution of the set of statistical distributions;

determining a second mean of a second statistical distribution of the set of statistical distributions; and

estimating the resource usage based on the first mean and the second mean.

8. The system of claim 6 , wherein the instructions are further executable by the processor for causing the processor to estimate the resource usage associated with the particular computing cluster by:

determining a first predefined percentile of a first statistical distribution of the set of statistical distributions;

determining a second predefined percentile of a second statistical distribution of the set of statistical distributions; and

estimating the resource usage based on the first predefined percentile and the second predefined percentile.

9. The system of claim 1 , wherein a statistical distribution of the set of statistical distributions corresponds to a particular software component, the statistical distribution including a first data value indicating a first number of times in which a first number of replicas of the particular software component concurrently executed in the particular computing cluster during the prior timespan, and the statistical distribution including a second data value indicating a second number of times in which a second number of replicas of the particular software component concurrently executed in the particular computing cluster during the prior timespan.

10. A method comprising:

receiving, by a processor, a request from a workload manager for identifying a computing cluster to which to assign a workload;

receiving, by the processor, historical information from a particular computing cluster, the historical information including a set of statistical distributions associated with a set of software components executed by the particular computing cluster during a prior timespan, wherein each statistical distribution in the set of statistical distributions indicates how many times the particular computing cluster executed different numbers of replicas of an associated software component during the prior timespan;

determining, by the processor, that the workload is to be assigned to the particular computing cluster among a plurality of computing clusters based on a metric derived from the set of statistical distributions in the historical information; and

transmitting, by the processor, a response to the workload manager for causing the workload manager to assign the workload to the particular computing cluster.

11. The method of claim 10 , wherein the historical information indicates statistical information about replicas of a software component deployed by the particular computing cluster during the prior timespan.

12. The method of claim 11 , wherein the software component is a serverless function or a microservice.

13. The method of claim 10 , wherein the plurality of computing clusters are edge clusters at physical edges of a distributed computing environment.

14. The method of claim 10 , further comprising:

estimating, by the processor, resource usage by the particular computing cluster based on the workload and the metric; and

selecting, by the processor, the particular computing cluster from among the plurality of computing clusters in response to determining that the estimated resource usage is less than a resource-usage threshold for the particular computing cluster.

15. The method of claim 14 , further comprising estimating the resource usage associated with the particular computing cluster by:

determining a first mean of a first statistical distribution of the set of statistical distributions;

determining a second mean of a second statistical distribution of the set of statistical distributions; and

estimating the resource usage based on the first mean and the second mean.

16. The method of claim 14 , further comprising estimating the resource usage associated with the particular computing cluster by:

determining a first predefined percentile of a first statistical distribution of the set of statistical distributions;

determining a second predefined percentile of a second statistical distribution of the set of statistical distributions; and

estimating the resource usage based on the first predefined percentile and the second predefined percentile.

17. The method of claim 10 , wherein a statistical distribution of the set of statistical distributions corresponds to a particular software component, the statistical distribution including a first data value indicating a first number of times in which a first number of replicas of the particular software component concurrently executed in the particular computing cluster during the prior timespan, and the statistical distribution including a second data value indicating a second number of times in which a second number of replicas of the particular software component concurrently executed in the particular computing cluster during the prior timespan.

18. A non-transitory computer-readable medium comprising program code that is executable by a processor for causing the processor to:

receive a request from a workload manager for identifying a computing cluster to which to assign a workload;

receive historical information from a particular computing cluster, the historical information including a set of statistical distributions associated with a set of software components executed by the particular computing cluster during a prior timespan, wherein each statistical distribution in the set of statistical distributions indicates how many times the particular computing cluster executed different numbers of replicas of an associated software component during the prior timespan;

determine that the workload is to be assigned to the particular computing cluster among a plurality of computing clusters based on a metric derived from the set of statistical distributions in the historical information; and

transmit a response to the workload manager for causing the workload manager to assign the workload to the particular computing cluster.

19. The non-transitory computer-readable medium of claim 18 , further comprising program code that is executable by the processor for causing the processor to:

estimate resource usage by the particular computing cluster based on the workload and the metric; and

select the particular computing cluster from among the plurality of computing clusters in response to determining that the estimated resource usage is less than a resource-usage threshold for the particular computing cluster.

20. The non-transitory computer-readable medium of claim 18 , wherein a statistical distribution of the set of statistical distributions corresponds to a particular software component, the statistical distribution including a first data value indicating a first number of times in which a first number of replicas of the particular software component concurrently executed in the particular computing cluster during the prior timespan, and the statistical distribution including a second data value indicating a second number of times in which a second number of replicas of the particular software component concurrently executed in the particular computing cluster during the prior timespan.