Simulating hosted application performance

Patent metadata

Field	Value
Publication number	US-10565083-B2
Application number	US-201715835977-A
Country	US
Kind code	B2
Filing date	Dec 8, 2017
Priority date	Dec 8, 2017
Publication date	Feb 18, 2020
Grant date	Feb 18, 2020

Abstract

A computer-implemented method including causing an application to execute on a private cloud computing network, collecting first performance metrics associated with the application as a result of the application executing on the private cloud computing network, generating a simulated workload based on the first performance metrics, causing the simulated workload to execute on one or more public cloud computing networks, collecting second performance metrics associated with the simulated workload as the simulated workload is executing on the one or more public clouds, and generating, based on the second performance metrics, a recommendation of one of the one or more public cloud computing networks to host the application is disclosed.

First claim

What is claimed is:

1. A computer-implemented method comprising:
at a server configured to connect in a network for communication with at least a private cloud computing network and one or more public cloud computing networks,
causing an application to execute on the private cloud computing network;
collecting first performance metrics associated with the application as a result of the application executing on the private cloud computing network;
generating a simulated workload based on the first performance metrics;
causing the simulated workload to execute on the one or more public cloud computing networks;
collecting second performance metrics associated with the simulated workload as the simulated workload is executing on the one or more public cloud computing networks; and
generating, based on the second performance metrics, a recommendation of one of the one or more public cloud computing networks to host the application.

2. The computer-implemented method of claim 1 , wherein the recommendation is based on one of a set of user profiles.

3. The computer-implemented method of claim 1 , wherein the generating further comprises:
after the application is deployed on one of the public cloud computing networks, obtaining updated performance metrics regarding the application as it is executing on one of the public cloud computing networks; and
updating a decision-making process for the recommendation based on the updated performance metrics.

4. The computer-implemented method of claim 3 , wherein the updating further comprises:
performing machine learning using the updated performance metrics.

5. The computer-implemented method of claim 4 , wherein the machine learning is performed on updated performance metrics from multiple executions of the application after it is deployed on the one or more public cloud computing networks.

6. The computer-implemented method of claim 1 , wherein executing the application on the private cloud computing network further comprises:
dynamically generating payload traffic to simulate a variety of network traffic loads.

7. The computer-implemented method of claim 1 , wherein the first and second performance metrics include at least one web server metric, wherein the web server metric is at least one of busy and idle threads, throughput, or bandwidth requirements.

8. The computer-implemented method of claim 1 , wherein the first and second performance metrics include at least one application server metric, and wherein the application server metric is:
at least one of load distribution, central processing unit (CPU) hotspots, or worker threads; or
at least one of time spent in a logical tier or a number of calls into a logical tier.

9. The computer-implemented method of claim 1 , wherein the server further comprises a communication interface, and wherein causing the simulated workload to execute on the one or more public cloud computing networks further comprises:
providing the simulated workload over the communication interface to the one or more public cloud computing networks to execute on the one or more public cloud computing networks.

10. An apparatus comprising:
a communication interface configured to enable network communications;
a processor coupled with the communication interface to communicate with at least a private cloud computing network and one or more public cloud computing networks, the processor configured to:
cause an application to execute on the private cloud computing network;
collect via the communication interface first performance metrics associated with the application as a result of the application executing on the private cloud computing network;
generate a simulated workload based on the first performance metrics;
provide the simulated workload via the communication interface to the one or more public cloud computing networks to execute on the one or more public cloud computing networks;
collect via the communication interface second performance metrics associated with the simulated workload as the simulated workload is executing on the one or more public cloud computing networks; and
generate, based on the second performance metrics, a recommendation of one of the one or more public cloud computing networks to host the application.

11. The apparatus of claim 10 , wherein the recommendation is based on one of a set of user profiles.

12. The apparatus of claim 10 , wherein the processor is further configured to:
after the application is deployed on one of the public cloud computing networks, obtain updated performance metrics regarding the application as it is executing on one of the public cloud computing networks; and
updating a decision-making process for the recommendation based on the updated performance metrics.

13. The apparatus of claim 12 , wherein the processor is further configured to:
perform machine learning using the updated performance metrics.

14. The apparatus of claim 13 , wherein the machine learning is performed on updated performance metrics from multiple executions of the application after it is deployed on the one or more public cloud computing networks.

15. The apparatus of claim 10 , wherein when the processor executes the application on the private cloud computing network, the processor is further configured to:
dynamically generate payload traffic to simulate a variety of network traffic loads.

16. The apparatus of claim 10 , wherein the first and second performance metrics include at least one web server metric, wherein the web server metric is at least one of busy and idle threads, throughput, or bandwidth requirements.

17. The apparatus of claim 10 , wherein the first and second performance metrics include at least one application server metric, wherein the application server metric is at least one of load distribution, central processing unit (CPU) hotspots, or worker threads.

18. The apparatus of claim 10 , wherein the first and second performance metrics include at least one application metric, wherein the application metric is at least one of time spent in a logical tier or a number of calls into a logical tier.

19. A computer program product comprising:
one or more non-transitory computer readable storage media;
instructions encoded on the non-transitory computer readable storage media;
the instructions being executable by a processor of a server configured to connect in a network for communications in order to:
cause an application to execute on a private cloud computing network;
collect first performance metrics associated with the application as a result of the application executing on the private cloud computing network;
generate a simulated workload based on the first performance metrics;
cause the simulated workload to execute on one or more public cloud computing networks;
collect second performance metrics associated with the simulated workload as the simulated workload is executing on the one or more public cloud computing networks; and
generate, based on the second performance metrics, a recommendation of one of the one or more public cloud computing networks to host the application.

20. The computer program product of claim 19 , wherein the recommendation is based on one of a set of user profiles.

Assignees

• Cisco Tech Inc

Inventors

• Mital Subhav
• Krattiger Lukas
• Kapadia Shyam N

Classifications

• G06F11/3428

  Benchmarking · CPC title

• G06F11/3457

  Performance evaluation by simulation · CPC title

• H04L43/0888

  Throughput · CPC title

• H04L67/306

  User profiles · CPC title

• H04L67/10

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title