Workload analyzer and optimizer integration

• US 6,622,221 B1
• Filed: 09/15/2000
• Issued: 09/16/2003
• Est. Priority Date: 08/17/2000
• Status: Expired due to Term

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Zahavi, William Z.
• Primary Examiner(s): Nguyen, Hiep T.
• Application Number:
  US09/663,540
• Time in Patent Office: 1,096 Days
• Field of Search: 711/114, 711/154, 711/170, 711/165, 703/2, 703/21, 703/22
• US Class Current: 711/154
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3447 Performance evaluation by modeling
• CPC Class Codes: G06F 11/3476 Data logging G06F11/14, G06F11/2205 take precedence
• CPC Class Codes: G06F 11/3485 for I/O devices
• CPC Class Codes: G06F 2003/0697 device management, e.g. handlers, drivers, I/O schedulers
• CPC Class Codes: G06F 3/0601 Interfaces specially adapted for storage systems
• CPC Class Codes: G06F 3/0611 in relation to response time
• CPC Class Codes: G06F 3/0653 Monitoring storage devices or systems
• CPC Class Codes: G06F 3/0683 Plurality of storage devices

Abstract

Described are techniques used in connection with evaluating the performance of proposed swap plans in accordance with past performance data collected. Data is gathered for a first arrangement of a data storage devices for a period of time. A proposed modification to the first arrangement, such as swapping logical volumes of data on one or more storage devices, is produced. A performance estimation is produced using a modeling technique, past performance data, and the proposed modification. The proposed modification may be produced using an automated tool, such as an optimizer, or other input means, such as by a user viewing actual performance data to determine obvious LV swap candidates. The proposed modification may be rejected, or accepted and implemented subsequent to viewing the performance estimation.

Claims

1. A method executed in a computer system for evaluating performance of at least one data storage device comprising:

gathering actual performance data associated with the at least one data storage device;

determining a modification in accordance with performance of the at least one data storage device;

modeling estimated performance of the at least one data storage device in connection with said modification and said actual performance data; and

determining, in response to modeling, whether to implement the modification.

2. The method of claim 1, further comprising:

viewing a portion of the actual performance data; and

determining, in response to viewing, said modification.

3. The method of claim 2, further comprising:

determining whether to implement the modification in accordance with user input.

4. The method of claim 1, further comprising:

determining the modification in connection with an automated technique included in a software tool.

5. The method of claim 4, further comprising:

executing a first portion of machine executable code to determine said modification.

6. The method of claim 5, wherein the first portion of machine executable code is included in an optimizer, and said executing of the first portion is performed in a data storage system including said at least one data storage device.

7. The method of claim 6, wherein the portion of actual performance data is viewed via a user interface.

8. The method of claim 6, further comprising:

specifying at least two portions of data included in said at least one data storage device to be swapped, and wherein said modification is a swap plan specifying said at least two portions of data.

9. The method of claim 8, wherein each of said at least two portions of data are logical volumes of disk data storage.

10. The method of claim 9, wherein said modeling estimated performance further comprises:

determining estimated service time for a data operation request in connection with said at least one data storage device.

11. The method of claim 10, further comprising:

determining estimated service time in connection with a cache miss of the at least one data storage device.

12. The method of claim 11, wherein there are at least two data storage devices, and the method further comprising:

gathering, at each of said at least two data storage devices, a corresponding portion of said actual performance data associated with performance of said each data storage device; and

collecting at least two corresponding portions of said actual performance data associated with said at least two data storage devices.

13. The method of claim 11, further comprising:

determining estimated response time for a data operation request in connection with said at least one data storage device.

14. A computer readable medium encoded for evaluating performance of at least one data storage device comprising:

machine executable code for gathering actual performance data associated with the at least one data storage device;

machine executable code for determining a modification in accordance with performance of the at least one data storage device;

machine executable code for modeling estimated performance of the at least one data storage device in connection with said modification and said actual performance data; and

machine executable code for determining whether to implement the modification.

15. The computer readable medium of claim 14, further comprising:

machine executable code for viewing a portion of the actual performance data; and

machine executable code for determining said modification.

16. The computer readable medium of claim 15, further comprising:

machine executable code for determining whether to implement the modification in accordance with user input.

17. The computer readable medium of claim 14, further comprising:

a first portion of machine executable code for determining the modification.

18. The computer readable medium of claim 17, wherein the first portion of machine executable code is included in an optimizer, and the computer system further includes a data storage system that includes said optimizer and said at least one data storage device.

19. The computer readable medium of claim 18, wherein the portion of actual performance data is viewed via a user interface.

20. The computer readable medium of claim 18, further comprising:

machine executable code for specifying at least two portions of data included in said at least one data storage device to be swapped, and wherein said modification is a swap plan specifying said at least two portions of data.

21. The computer readable medium of claim 20, wherein each of said at least two portions of data are logical volumes of disk data storage.

22. The computer readable medium of claim 21, wherein said machine executable code for modeling estimated performance further comprises:

machine executable code for determining estimated service time for a data operation request in connection with said at least one data storage device.

23. The computer readable medium of claim 22, further comprising:

machine executable code for determining estimated service time in connection with a cache miss of the at least one data storage device.

24. The computer readable medium of claim 23, wherein there are at least two data storage devices, and the computer readable medium further comprising:

machine executable code for gathering, at each of said at least two data storage devices, a corresponding portion of said actual performance data associated with performance of said each data storage device; and

machine executable code for collecting at least two corresponding portions of said actual performance data associated with said at least two data storage devices.

25. The computer readable medium of claim 23, further comprising:

machine executable code for determining estimated response time for a data operation request in connection with said at least one data storage device.

26. A computer system comprising:

a computer system resource used by at least one host;

a work analyzer for performing performance analysis of the computer system resource in accordance with actual performance data;

a modeler for estimating performance of the computer system resource in accordance with a modification associated with the computer system resource and a portion of the actual performance data; and

an evaluator, coupled to the modeler, for determining whether to implement said modification.

27. The computer system of claim 26, further comprising:

a data storage system including at least one data storage device, wherein the computer system resource is the at least one data storage device.

28. The computer system of claim 27, further comprising:

an optimizer producing a swap plan in connection with performance of the at least one data storage device, wherein the swap plan is said modification.

29. The computer system of claim 28, further comprising:

a user interface for viewing the performance analysis;

user input, coupled to the evaluator and the user interface, for determining whether to implement the modification.

30. The computer system of claim 29, wherein said at least one host system is connected to the data storage system, the optimizer is included in the data storage system, the work analyzer, the user interface, the modeler, and the evaluator are included in data manager system.

31. The computer system of claim 30, further comprising:

an agent associated with said computer system resource for gathering actual performance data in connection with said computer system resource.

32. The computer system of claim 31, further comprising:

a data collector for collecting actual performance data produced by the agent.

33. The computer system of claim 32, further comprising:

an archive of actual performance data associated with a predetermined period of time, wherein the portion of actual performance data used by the modeler is included in said archive.

34. The computer system of claim 33, wherein the actual performance data includes base parameters.

35. The computer system of claim 34, wherein the actual performance data includes derived parameters produced using at least one base parameter.

36. A method executed in a computer system for performing performance analysis comprising:

performing data operations in connection with a computer system resource by at least one host system;

performing, by a work analyzer, performance analysis of the computer system resource in accordance with actual performance data;

estimating, by a modeler, performance of the computer system resource in accordance with a modification associated with the computer system resource and a portion of the actual performance data; and

determining, by an evaluator, whether to implement said modification.

37. The method of claim 36, wherein the computer system includes a data storage system including at least one data storage device, wherein the computer system resource is the at least one data storage device.

38. The method of claim 37, further comprising:

producing, using an optimizer, a swap plan in connection with performance of the at least one data storage device, wherein the swap plan is said modification.

39. The method of claim 38, further comprising:

viewing the performance analysis with a user interface;

determining, in accordance with user input, whether to implement the modification.

40. The method of claim 39, wherein said at least one host system is connected to the data storage system, the optimizer is included in the data storage system, the work analyzer, the user interface, and the modeler are included in a data manager system.

41. The method of claim 40, further comprising:

gathering actual performance data in connection with said computer system resource by an agent associated with the computer system resource.

42. The method of claim 41, further comprising:

collecting, by a data collector, actual performance data produced by the agent.

43. The method of claim 42, further comprising:

archiving actual performance data associated with a predetermined period of time, wherein the portion of actual performance data used by the modeler is included in said archive.

44. The method of claim 43, wherein the actual performance data includes base parameters.

45. The method of claim 44, wherein the actual performance data includes at least one derived parameter produced using at least one base parameter.

46. A method executed in a computer system for evaluating performance of at least one data storage device comprising:

gathering actual performance data associated with the at least one data storage device;

determining a modification in accordance with performance of the at least one data storage device;

modeling estimated performance of the at least one data storage device in connection with said modification and said actual performance data;

presenting said estimated performance in connection with said modification in a visual form to a user; and

determining, in response to presenting, whether to implement the modification.

47. The method of claim 46, wherein said visual form is one of a tabular form and a graphical form.

48. The method of claim 47, further comprising:

receiving said response using a graphical user interface; and

implementing the modification by an optimizer associated with the at least one data storage device.

49. The method of claim 48, further comprising:

specifying at least two portions of data included in said at least one data storage device to be swapped, and wherein the modification is a swap plan specifying said at least two portions.

50. The method of claim 49, wherein each of said at least two portions of data are logical volumes of disk data storage.

51. The method of claim 46, further comprising:

presenting a visual display in connection with performance of said at least one data storage device without said modification and the actual performance data, and wherein said determining whether to implement said modification is also performed in response to said presenting a visual display in connection with performance of said at least one data storage device without said modification.

52. A computer readable medium encoded for evaluating performance of at least one data storage device comprising:

machine executable code for gathering actual performance data associated with the at least one data storage device;

machine executable code for determining a modification in accordance with performance of the at least one data storage device;

machine executable code for modeling estimated performance of the at least one data storage device in connection with said modification and said actual performance data;

machine executable code for presenting said estimated performance in connection with said modification in a visual form to a user; and

machine executable code for determining whether to implement the modification which executes in response to said machine executable code for presenting said estimated performance.

53. The computer readable medium of claim 52, wherein said visual form is one of a tabular form and a graphical form.

54. The computer readable medium of claim 53, further comprising:

machine executable code for receiving said response using a graphical user interface; and

machine executable code for implementing the modification by an optimizer associated with the at least one data storage device.

55. The computer readable medium of claim 54, further comprising:

machine executable code for specifying at least two portions of data included in said at least one data storage device to be swapped, and wherein the modification is a swap plan specifying said at least two portions.

56. The computer readable medium of claim 55, wherein each of said at least two portions of data are logical volumes of disk data storage.

57. The computer readable medium of claim 52, further comprising:

machine executable code for presenting a visual display in connection with performance of said at least one data storage device without said modification and the actual performance data, and wherein said machine executable code for determining whether to implement said modification is executed in response to said machine executable code for presenting a visual display in connection with performance of said at least one data storage device without said modification.

58. A method executed in a computer system for determining whether to implement a swap plan comprising:

performing data operations in connection with at least one data storage device included in said computer system;

collecting performance data in connection with said data operations;

determining a swap plan in accordance with a first portion of performance data associated with said data operations;

estimating, using a second portion of said performance data, a first response time associated with said at least one data storage device without implementing said swap plan;

estimating, using said second portion of said performance data, a second response time associated with said at least one data storage device with said swap plan; and

determining whether to implement said swap plan in accordance with said first and said second response times.

59. The method of claim 58, wherein said swap plan includes a first logical volume and a second logical volume each associated with a different data storage device, and the method further including:

swapping a first set of data associated with said first logical volume with a second set of data associated with said second logical volume if it is determined to implement said swap plan.

60. The method of claim 58, further comprising:

analyzing said first portion of said performance data using an optimizer; and

determining, in response to analyzing, said swap plan using said optimizer.

61. The method of claim 58, wherein said estimating said first and second response times includes:

modeling using said second portion of performance data associated with a plurality of components associated with said at least one data storage device, and

wherein determining a swap plan includes;

analyzing said first portion of performance data associated with at least one of said plurality of components.

62. The method of claim 61, wherein at least one common performance data metric is included in both said first portion and said second portion of performance data.

63. A computer readable medium encoded for determining whether to implement a swap plan comprising:

machine executable code for performing data operations in connection with at least one data storage device included in said computer system;

machine executable code for collecting performance data in connection with said data operations;

machine executable code for determining a swap plan in accordance with a first portion of performance data associated with said data operations;

machine executable code for estimating, using a second portion of said performance data, a first response time associated with said at least one data storage device without implementing said swap plan;

machine executable code for estimating, using said second portion of said performance data, a second response time associated with said at least one data storage device with said swap plan; and

machine executable code for determining whether to implement said swap plan in accordance with said first and said second response times.

64. The computer readable medium of claim 63, wherein said swap plan includes a first logical volume and a second logical volume each associated with a different data storage device, and the computer readable medium further including:

machine executable code for swapping a first set of data associated with said first logical volume with a second set of data associated with said second logical volume if it is determined to implement said swap plan.

65. The computer readable medium of claim 63, further comprising:

machine executable code for analyzing said first portion of said performance data using an optimizer; and

machine executable code for determining said swap plan using said optimizer.

66. The computer readable medium of claim 63, wherein said machine executable code for estimating said first and second response times includes:

machine executable code for modeling using said second portion of performance data associated with a plurality of components associated with said at least one data storage device, and

wherein said machine executable code for determining a swap plan includes;

machine executable code for analyzing said first portion of performance data associated with at least one of said plurality of components.

67. The computer readable medium of claim 66, wherein at least one common performance data metric is included in both said first portion and said second portion of performance data.

68. A method for evaluating performance of at least one data storage device comprising:

determining a modification in accordance with performance of the at least one data storage device;

modeling estimated performance of the at least one data storage device in connection with said modification and actual performance data associated with said at least one data storage device; and

determining, in response to said modeling, whether to implement the modification.

69. The method of claim 68, further comprising:

performing at least one data operation using said at least one data storage device; and

performing a performance analysis of said at least one data storage device in accordance with said at least one data operation.

70. The method of claim 69, wherein the at least one data storage device is included in a computer system, and the method further comprising:

producing a swap plan in connection with performance of said at least one data storage system wherein the swap plan includes the modification, said swap plan indicating at least two portions of data of said at least one data storage device to be swapped.

71. The method of claim 70, further comprising:

determining the modification using an optimizer.

72. The method of claim 68, further comprising:

gathering performance data about said at least one data storage device.

73. The method of claim 72, further comprising:

storing said performance data, wherein said performance data includes at least one base parameter and at least one derived parameter produced using said at least one base parameter.

74. The method of claim 68, further comprising:

displaying one of a graphical representation and a tabular representation of said estimated performance;

receiving a user input in response to said display; and

determining whether to implement said modification in response to said user input.

75. A computer readable medium encoded for evaluating performance of at least one data storage device comprising:

machine executable instructions that determine a modification in accordance with performance of the at least one data storage device;

machine executable instructions that model estimated performance of the at least one data storage device in connection with said modification and actual performance data associated with said at least one data storage device; and

machine executable instructions that determine whether to implement the modification.

76. The computer readable medium of claim 75, further comprising:

machine executable instructions that perform at least one data operation using said at least one data storage device; and

machine executable instructions that perform a performance analysis of said at least one data storage device in accordance with said at least one data operation.

77. The computer readable medium of claim 76, wherein the at least one data storage device is included in a computer system, and the computer readable medium further comprising:

machine executable instructions that produce a swap plan in connection with performance of said at least one data storage system wherein the swap plan includes the modification, said swap plan indicating at least two portions of data of said at least one data storage device to be swapped.

78. The computer readable medium of claim 77, further comprising:

machine executable instructions that determine the modification using an optimizer.

79. The computer readable medium of claim 75, further comprising:

machine executable instructions that gather performance data about said at least one data storage device.

80. The computer readable medium of claim 77, further comprising:

machine executable instructions that store said performance data, wherein said performance data includes at least one base parameter and at least one derived parameter produced using said at least one base parameter.

81. The computer readable medium of claim 75, further comprising:

machine executable instructions that display one of a graphical representation and a tabular representation of said estimated performance machine executable instructions that receive a user input in response to said display; and

machine executable instructions that determine whether to implement said modification in response to said user input.