Allocation of shared computing resources using source code feature extraction and machine learning

Patent metadata

Field	Value
Publication number	US-11567807-B2
Application number	US-201815941434-A
Country	US
Kind code	B2
Filing date	Mar 30, 2018
Priority date	Mar 30, 2018
Publication date	Jan 31, 2023
Grant date	Jan 31, 2023

Abstract

Techniques are provided for allocation of shared computing resources using source code feature extraction and machine learning techniques. An exemplary method comprises obtaining source code for execution in a shared computing environment; extracting a plurality of discriminative features from the source code; obtaining a trained machine learning model; and generating a prediction of an allocation of one or more resources of the shared computing environment needed to satisfy one or more service level agreement requirements for the source code. The generated prediction is optionally adjusted using a statistical analysis of an error curve, based on one or more error boundaries obtained by the trained machine learning model. The trained machine learning model can be trained using a set of discriminative features extracted from training source code and corresponding measurements of metrics of the service level agreement requirements obtained by executing the training source code on a plurality of the resources of the shared computing environment.

First claim

What is claimed is:

1. A method, comprising:
obtaining source code for execution in a shared computing environment;
extracting, using at least one processing device, a plurality of discriminative features from the source code;
obtaining, using the at least one processing device, at least one trained machine learning model; and
applying, using the at least one processing device, the plurality of extracted discriminative features to the at least one trained machine learning model, wherein the at least one trained machine learning model generates a prediction of an allocation comprising a required amount of one or more resources of the shared computing environment needed to satisfy one or more service level agreement requirements for the source code to be executed in the shared computing environment, and wherein the at least one trained machine learning model is trained using a set of discriminative features extracted from training source code and corresponding measurements of one or more metrics of the service level agreement requirements obtained by executing the training source code on a plurality of the resources of the shared computing environment.

2. The method of claim 1 , further comprising generating one or more error boundaries of the predicted allocation of the one or more resources based on a standard deviation of the at least one machine learning model.

3. The method of claim 1 , wherein the extracting the plurality of discriminative features from the source code employs one or more of natural language processing techniques, pattern-based techniques and worst-case execution time techniques.

4. The method of claim 3 , wherein the natural language processing techniques extract one or more of the discriminative features from the source code based on a term frequency metric.

5. The method of claim 3 , wherein the pattern-based techniques extract one or more of the discriminative features from the source code by identifying one or more patterns in function calls of the source code.

6. The method of claim 3 , wherein the worst-case execution time techniques extract one or more of the discriminative features from the source code by obtaining statistics regarding execution times for each type of task in the source code.

7. The method of claim 1 , further comprising estimating an amount of time required to process the source code until a completion of the execution of the source code.

8. A system, comprising:
a memory; and
at least one processing device, coupled to the memory, operative to implement the following steps:
obtaining source code for execution in a shared computing environment;
extracting, using at least one processing device, a plurality of discriminative features from the source code;
obtaining, using the at least one processing device, at least one trained machine learning model; and
applying, using the at least one processing device, the plurality of extracted discriminative features to the at least one trained machine learning model, wherein the at least one trained machine learning model generates a prediction of an allocation comprising a required amount of one or more resources of the shared computing environment needed to satisfy one or more service level agreement requirements for the source code to be executed in the shared computing environment, and wherein the at least one trained machine learning model is trained using a set of discriminative features extracted from training source code and corresponding measurements of one or more metrics of the service level agreement requirements obtained by executing the training source code on a plurality of the resources of the shared computing environment.

9. The system of claim 8 , further comprising the step of generating one or more error boundaries of the predicted allocation of the one or more resources based on a standard deviation of the at least one machine learning model.

10. The system of claim 8 , wherein the step of extracting the plurality of discriminative features from the source code employs one or more of natural language processing techniques, pattern-based techniques and worst-case execution time techniques.

11. The system of claim 10 , wherein the natural language processing techniques extract one or more of the discriminative features from the source code based on a term frequency metric.

12. The system of claim 10 , wherein the pattern-based techniques extract one or more of the discriminative features from the source code by identifying one or more patterns in function calls of the source code.

13. The system of claim 10 , wherein the worst-case execution time techniques extract one or more of the discriminative features from the source code by obtaining statistics regarding execution times for each type of task in the source code.

14. The system of claim 8 , further comprising estimating an amount of time required to process the source code until a completion of the execution of the source code.

15. A computer program product, comprising a tangible machine-readable storage medium having encoded therein executable code of one or more software programs, wherein the one or more software programs when executed by at least one processing device perform the following steps:
obtaining source code for execution in a shared computing environment;
extracting, using at least one processing device, a plurality of discriminative features from the source code;
obtaining, using the at least one processing device, at least one trained machine learning model; and
applying, using the at least one processing device, the plurality of extracted discriminative features to the at least one trained machine learning model, wherein the at least one trained machine learning model generates a prediction of an allocation comprising a required amount of one or more resources of the shared computing environment needed to satisfy one or more service level agreement requirements for the source code to be executed in the shared computing environment, and wherein the at least one trained machine learning model is trained using a set of discriminative features extracted from training source code and corresponding measurements of one or more metrics of the service level agreement requirements obtained by executing the training source code on a plurality of the resources of the shared computing environment.

16. The computer program product of claim 15 , further comprising the step of generating one or more error boundaries of the predicted allocation of the one or more resources based on a standard deviation of the at least one machine learning model.

17. The computer program product of claim 15 , wherein the step of extracting the plurality of discriminative features from the source code employs one or more of natural language processing techniques, pattern-based techniques and worst-case execution time techniques.

18. The computer program product of claim 17 , wherein the natural language processing techniques extract one or more of the discriminative features from the source code based on a term frequency metric.

19. The computer program product of claim 17 , wherein the pattern-based techniques extract one or more of the discriminative features from the source code by identifying one or more patterns in function calls of the source code.

20. The computer program product of claim 15 , further comprising estimating an amount of time required to process the source code until a completion of the execution of the source code.

Assignees

• Emc Ip Holding Co Llc

Inventors

• Dias Jonas F
• Calmon Tiago Salviano
• Prado Adriana Bechara

Classifications

• G06F8/433

  Dependency analysis; Data or control flow analysis · CPC title

• H04L47/2425

  for supporting services specification, e.g. SLA · CPC title

• G06N20/00

  Machine learning · CPC title

• G06F40/40

  Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30) · CPC title

• H04L47/823

  Electricity · mapped topic