Calibrating cloud computing environments

• US 9,323,561 B2
• Filed: 08/13/2010
• Issued: 04/26/2016
• Est. Priority Date: 08/13/2010
• Status: Active Grant

• Current Assignee: Kyndryl Incorporated
• Original Assignee: International Business Machines Corporation
• Inventors: Ayala, Richard, Chavda, Kavita, Gopisetty, Sandeep, Murthy, Seshashayee S., Singh, Aameek
• Primary Examiner(s): Chang, Tom Y
• Application Number:
  US12/855,780
• Publication Number:
  US 20120042061A1
• Time in Patent Office: 2,083 Days
• Field of Search: 709/224
• US Class Current: 1/1
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3433 for load management allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 2209/501 Performance criteria
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: H04L 41/14 Network analysis or design
• CPC Class Codes: H04L 41/145 involving simulating, designing, planning or modelling of a network

Abstract

In general, embodiments of present invention provide an approach for calibrating a cloud computing environment. Specifically, embodiments of the present invention provide an empirical approach for obtaining end-to-end performance characteristics for workloads in the cloud computing environment (hereinafter the “environment”). In a typical embodiment, different combinations of cloud server(s) and cloud storage unit(s) are determined. Then, a virtual machine is deployed to one or more of the servers within the cloud computing environment. The virtual machine is used to generate a desired workload on a set of servers within the environment. Thereafter, performance measurements for each of the different combinations under the desired workload will be taken. Among other things, the performance measurements indicate a connection quality between the set of servers and the set of storage units, and are used in calibrating the cloud computing environment to determine future workload placement. Along these lines, the performance measurements can be populated into a table or the like, and a dynamic map of a data center having the set of storage units can be generated.

Claims

1. A method for calibrating performance within a cloud computing environment, comprising:

determining, using at least one computing device, different server-storage unit pair combinations of representative devices from at least one cluster of a set of servers and at least one cluster of a set of storage units within the cloud computing environment, the representative devices comprising a same hardware characteristic and a similar configuration to other devices of the at least one clusters, the same hardware characteristic comprising a connectivity with storage or a network;

generating, using the at least one computing device, a desired workload on the representative device combinations,

wherein the workload is created by a virtual machine, flash copied, and deployed to a server of each representative device combination, and

wherein the desired workload is configured to call for storage of data in a storage unit of each representative device combination via a server of said representative device combination;

taking, using the at least one computing device, performance measurements with an application of the virtual machine for each representative server-storage unit pair device combination under the desired workload;

calculating a connection quality between the storage unit and the server of each representative device combination based on the performance measurements;

assigning, for each representative device combination of a server and a storage unit, the calculated connection quality to combinations of a server from a cluster of servers and a storage unit from a cluster of storage units, wherein the clusters are associated with the representative device combination;

determining a plurality of optimal server-storage unit pairings from among the at least one cluster of servers and the at least one cluster of storage units based on the assigned qualities of connection;

calibrating the plurality of optimal server-storage unit pairings in the cloud computing environment for an anticipated workload placement; and

balancing at least one workload among at least one optimized combination of a server from the at least one cluster of a set of servers and a storage unit from the at least one cluster of a set of storage units within the cloud computing environment based on the assigned connection qualities and the calibration.

2. The method of claim 1, further comprising generating, using the at least one computing device, a table that comprises the performance measurements.

3. The method of claim 2, further comprising creating, using the at least one computing device, a dynamic map of a data center comprising the set of storage units based on the table.

4. The method of claim 1, the performance measurements indicating a connection quality comprising at least one of:

speed, bandwidth, error rate, and network latency, between the set of servers and the set of storage units.

5. The method of claim 4, further comprising calibrating, using the at least one computing device, the cloud computing environment using the performance metrics to determine future workload placement.

6. The method of claim 1, the virtual machine comprising:

a simulation and measurement application;

an operating system; and

a resource definition library.

7. The method of claim 1, the virtual machine being created, using the at least one computing device, using an image library comprising a set of images for a simulation and measurement application and an operating system, and comprising a resource definition library.

8. The method of claim 1, the workload created by the virtual machine being scaled and allotted a scaled storage based on a predetermined accuracy scale factor.

9. A system for calibrating performance within a cloud computing environment, comprising:

a bus;

a processor coupled to the bus; and

a memory medium coupled to the bus, the memory medium comprising instructions to;

group a set of servers and a set of storage units each into at least one cluster based on a same hardware characteristic and a similar configuration, the same hardware characteristic comprising a connectivity with storage or a network;

select at least one representative device from within each of the at least one clusters, the at least one representative device comprising a similarity to other devices of the at least one cluster;

determine different server-storage unit pair combinations of the set of servers and the set of storage units within the cloud computing environment;

generate a desired workload on at least one combination of the -representative devices,

wherein the workload is created by a virtual machine, flash copied, and deployed to a server of each representative device combination, and

wherein the desired workload is configured to call for storage of data in the storage unit of the combination of the representative devices via the server of the combination of the representative devices;

take performance measurements with an application of the virtual machine for the combination of the representative server-storage unit pair devices under the desired workload;

calculate a connection quality between the storage unit and the server of the combination of the representative devices based on the performance measurements;

assign the calculated connection quality to combinations of a server and a storage unit from the at least one clusters of servers and of storage units associated with the combination of representative devices;

determine a plurality of optimal server-storage unit pairings from among the at least one clusters of servers and of storage units based on the assigned qualities of connection;

calibrating the plurality of optimal server-storage unit pairings in the cloud computing environment for an anticipated workload placement; and

balance at least one workload among at least one optimized combination of a server from the set of servers and a storage unit from the set of storage units within the cloud computing environment based on the assigned connection qualities and the calibration.

10. The system of claim 9, the memory medium further comprising instructions to generate a table that comprises the performance measurements.

11. The system of claim 10, the memory medium further comprising instructions to create a dynamic map of a data center comprising the set of storage units based on the table.

12. The system of claim 9, the performance measurements indicating a connection quality comprising at least one of:

speed, bandwidth, error rate, and network latency, between the set of servers and the set of storage units.

13. The system of claim 12, the memory medium further comprising instructions to calibrate the cloud computing environment using the performance metrics to determine future workload placement.

14. The system of claim 9, the virtual machine comprising:

a simulation and measurement application;

an operating system; and

a resource definition library.

15. The system of claim 9, the virtual machine being created using an image library comprising a set of images for a simulation and measurement application and an operating system, and comprising a resource definition library.

16. The system of claim 9, the workload created by the virtual machine being scaled and allotted a scaled storage based on a predetermined accuracy scale factor.

17. A computer program product for calibrating a cloud computing environment, the computer program product comprising a computer readable storage media, and program instructions stored on the computer readable storage media, to:

determine different server-storage unit pair combinations of representative devices from at least one cluster of a set of servers and at least one cluster of a set of storage units within the cloud computing environment, the representative devices comprising a same hardware characteristic and a similar configuration to other devices of the at least one clusters, the same hardware characteristic comprising a connectivity with storage or a network;

generate a desired workload on the representative device combinations,

wherein the workload is created by a virtual machine, flash copied, and deployed to a server of each representative device combination, and

wherein the desired workload is configured to call for storage of data in a storage unit of each representative device combination via a server of said representative device combination;

take performance measurements with an application of the virtual machine for each representative device combination under the desired workload;

calculate a connection quality between the storage unit and the server of each representative device combination based on the performance measurements;

assign, for each representative device combination, the calculated connection quality to combinations of a server from a cluster of servers and a storage unit from a cluster of storage units, wherein the clusters are associated with the representative device combination;

determine a plurality of optimal server-storage unit pairings from among the at least one cluster of servers and the at least one cluster of storage units based on the assigned qualities of connection;

calibrate the plurality of optimal server-storage unit pairings in the cloud computing environment for an anticipated workload placement; and

balance at least one workload among at least one optimized combination of a server from the at least one cluster of a set of servers and a storage unit from the at least one cluster of a set of storage units within the cloud computing environment based on the assigned connection qualities and the calibration.

18. The computer program product of claim 17, further comprising program instructions stored on the computer readable storage media to generate a table that comprises the performance measurements.

19. The computer program product of claim 18, the memory medium further comprising instructions to create a dynamic map of a data center comprising the set of storage units based on the table.

20. The computer program product of claim 17, the performance measurements indicating a connection quality comprising at least one of:

speed, bandwidth, error rate, and network latency, between the set of servers and the set of storage units.

21. The computer program product of claim 20, the memory medium further comprising instructions to calibrate the cloud computing environment using the performance metrics to determine future workload placement.

22. The computer program product of claim 17, the virtual machine comprising:

a simulation and measurement application;

an operating computer program product; and

a resource definition library.

23. The computer program product of claim 17, the virtual machine being created using an image library comprising a set of images for a simulation and measurement application and an operating system, and comprising a resource definition library.

24. The computer program product of claim 17, the workload created by the virtual machine being scaled and allotted a scaled storage based on a predetermined accuracy scale factor.

25. A method for deploying a system for calibrating performance within a cloud computing environment, comprising:

providing a computer infrastructure being operable to;

group a set of servers and a set of storage units each into clusters based on a same hardware characteristic and a similar configuration, the same hardware characteristic comprising a connectivity with storage or a network;

select at least one representative device from within each of the clusters, the at least one representative device comprising a similarity to other devices of the respective cluster;

determine different server-storage unit pair combinations of the at least one representative device from within each of the clusters within the cloud computing environment;

generate a desired workload on at least one representative device combination from within each of the clusters,

wherein the workload is created by a virtual machine, flash copied, and deployed to a server of each representative device combination, and

wherein the desired workload is configured to call for storage of data in a storage unit of the representative server-storage unit pair devices combination via a server of the representative devices combination;

take performance measurements with an application of the virtual machine for the representative device combinations under the desired workload;

calculate a connection quality between a storage unit and a server of the representative device combination based on the performance measurements;

assign the calculated connection quality to combinations of a server and a storage unit from the clusters associated with the representative devices;

determine a plurality of optimal server-storage unit pairings from among the at least one clusters of servers and of storage units based on the assigned qualities of connection;

calibrating the plurality of optimal server-storage unit pairings in the cloud computing environment for an anticipated workload placement; and

balance at least one workload among at least one optimized combination of a server from the set of servers and a storage unit from the set of storage units within the cloud computing environment based on the assigned connection qualities and the calibration.