System and method for determining how many servers of at least one server configuration to be included at a service provider's site for supporting an expected workload

• US 20050138170A1
• Filed: 12/17/2003
• Published: 06/23/2005
• Est. Priority Date: 12/17/2003
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Tang, Wenting, Cherkasova, Ludmila
• Granted Patent: US 8,145,731 B2
• Time in Patent Office: Days
• US Class Current: 709/225
• CPC Class Codes: H04L 41/145 involving simulating, designing, planning or modelling of a network
• CPC Class Codes: H04L 41/5003 Managing SLA; Interaction between SLA and QoS
• CPC Class Codes: H04L 41/5006 Creating or negotiating SLA contracts, guarantees or penalties

Abstract

A method comprises receiving, into a capacity planning system, workload information representing an expected workload of client accesses of streaming media files from a site. The method further comprises the capacity planning system determining, for at least one server configuration, how many servers of the at least one server configuration to be included at the site for supporting the expected workload in a desired manner.

Claims

1. A method comprising:

receiving, into a capacity planning system, workload information representing an expected workload of client accesses of streaming media files from a site; and

said capacity planning system determining, for at least one server configuration, how many servers of said at least one server configuration to be included at said site for supporting the expected workload in a desired manner.

2. The method of claim 1 comprising:

said capacity planning system determining, for each of a plurality of different server configurations, how many servers of each server configuration to be included at said site for supporting the expected workload in a desired manner.

3. The method of claim 1 wherein said workload information includes identification of a number of concurrent client accesses of said streaming media files from said site over a period of time.

4. The method of claim 3 wherein said workload information further includes identification of a corresponding encoding bit rate of each of said streaming media files accessed.

5. The method of claim 1 wherein said workload information comprises information from an access log collected over a period of time.

6. The method of claim 1 further comprising:

receiving, into said capacity planning system, configuration information for said at least one server configuration.

7. The method of claim 6 wherein said configuration information includes identification of size of memory of said at least one server configuration.

8. The method of claim 1 wherein said determining how many servers of each configuration to be included at said site for supporting the workload in a desired manner comprises:

computing a cost corresponding to resources of said at least one server configuration that are consumed in supporting the workload.

9. The method of claim 8 wherein said computing said cost comprises:

computing a cost of consumed resources for a stream in said workload having a memory access to a streaming media file; and

computing a cost of consumed resources for a stream in said workload having a disk access to a streaming media file.

10. The method of claim 1 wherein said determining how many servers of each configuration to be included at said site for supporting the workload in a desired manner comprises:

computing a service demand for said at least one server configuration in supporting said expected workload.

11. The method of claim 10 wherein said computing said service demand comprises computing:

$\mathrm{Demand}=\underset{i=1}{\overset{K_W}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>N_{X_{W_i}}^{\mathrm{memory}}\times <br><br>{\mathrm{cost}}_{X_{W_i}}^{\mathrm{memory}}+\underset{i=1}{\overset{K_W}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>N_{X_{W_i}}^{\mathrm{disk}}\times <br><br>{\mathrm{cost}}_{X_{W_i}}^{\mathrm{disk}},$

wherein the workload W comprises

$X_w=X_1,\mathrm{\dots <br><br>}<br><br>\text{\hspace{1em}<br><br>},X_{k_w}$

set of different encoded bit rates of files served in the workload,

$N_{X_{w_i}}^{\mathrm{memory}}$

is a number of streams in the workload having a memory access to a subset of files encoded at

$X_{W_i}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{Kb}<br><br>\text{/}<br><br>s,{\mathrm{cost}}_{X_{W_i}}^{\mathrm{memory}}$

is a cost of consumed resources for a stream having a memory access to a file encoded at

$X_{W_i}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{Kb}<br><br>\text{/}<br><br>s,N_{X_{W_i}}^{\mathrm{disk}}$

is a number of streams in the workload having a disk access to a subset of files encoded at

$X_{W_i}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{Kb}<br><br>\text{/}<br><br>s,,\mathrm{and}<br><br>\text{\hspace{1em}<br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{cost}}_{X_{W_i}}^{\mathrm{disk}}$

is a cost of consumed resources for a stream having a disk access to a file encoded at

$X_{W_i}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{Kb}<br><br>\text{/}<br><br>s.$

12. The method of claim 11 further comprising:

said capacity planning system determining from said computed service demand how many servers of said at least one server configuration to be included at said site for supporting the expected workload in the desired manner.

13. The method of claim 11 further comprising:

said capacity planning system determining from said computed service demand whether a single server of said at least one server configuration can support the expected workload in the desired manner.

14. The method of claim 13 further comprising:

if said capacity planning system determines from said computed service demand that a single server of said at least one server configuration cannot support the expected workload in the desired manner, said capacity planning system estimating from said computed service demand a number of servers of said at least one server configuration that can support the expected workload in the desired manner.

15. The method of claim 14 further comprising:

said capacity planning system re-evaluating the expected workload for a cluster having the estimated number of servers of said at least one server configuration to compute a service demand for the cluster to determine whether the estimated number of servers is accurate.

16. The method of claim 15 wherein said re-evaluating includes considering a load balancing strategy used by said cluster in supporting the expected workload.

17. The method of claim 1 further comprising:

receiving at least one service parameter.

18. The method of claim 17 wherein said at least one service parameter comprises information identifying at least one performance criteria desired to be satisfied by said site in supporting the expected workload in the desired manner.

19. The method of claim 18 wherein said at least one performance criteria specifies a minimum percentage of time that said site is desired to be capable of supporting the workload.

20. The method of claim 17 wherein said at least one service parameter comprises information identifying a constraint.

21. The method of claim 17 wherein said determining how many servers of said at least one server configuration to be included at said site for supporting the expected workload in a desired manner comprises:

determining how many servers of said at least one server configuration to be included at said site for supporting the expected workload in a manner that satisfies said at least one service parameter.

22. The method of claim 1 wherein said determining how many servers of said at least one server configuration to be included at said site for supporting the expected workload in a desired manner comprises:

computing, for resources of a single server of said at least one server configuration, a first service demand under the expected workload;

determining from said first service demand if said single server has sufficient capacity for supporting the expected workload in the desired manner; and

if determined that said single server does not have sufficient capacity for supporting the expected workload in the desired manner, estimating from the first service demand a number of servers of said at least one server configuration that when clustered together provide sufficient capacity for supporting the expected workload in the desired manner, and computing, for resources of the cluster of the number of servers, a second service demand under the expected workload.

23. The method of claim 23 further comprising:

analyzing the second service demand to verify whether the number of servers estimated from the first service demand is accurate.

24. A method comprising:

receiving, into a capacity planning tool, information about a first server configuration;

receiving, into said capacity planning tool, workload information representing an expected workload of client accesses of streaming media files from a site; and

said capacity planning tool determining how many servers of said first server configuration to be included at said site for supporting the expected workload in a desired manner.

25. The method of claim 24 wherein said determining how many servers of said first server configuration to be included at said site for supporting the expected workload in a desired manner comprises:

computing a cost corresponding to resources of said first server configuration that are consumed in supporting the workload.

26. The method of claim 25 wherein said computing said cost comprises:

computing a cost of consumed resources for a stream in said workload having a memory access to a streaming media file; and

computing a cost of consumed resources for a stream in said workload having a disk access to a streaming media file.

27. The method of claim 25 wherein said determining how many servers of said first server configuration to be included at said site for supporting the expected workload in a desired manner comprises:

using the computed cost to compute a service demand for said first server configuration in supporting said expected workload.

28. The method of claim 27 wherein said computing said service demand comprises computing:

$\mathrm{Demand}=\underset{i=1}{\overset{K_W}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>N_{X_{W_i}}^{\mathrm{memory}}\times <br><br>{\mathrm{cost}}_{X_{W_i}}^{\mathrm{memory}}+\underset{i=1}{\overset{K_W}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>N_{X_{W_i}}^{\mathrm{disk}}\times <br><br>{\mathrm{cost}}_{X_{W_i}}^{\mathrm{disk}},$

wherein the workload W comprises

$X_w=X_1,\mathrm{\dots <br><br>},X_{k_w}$

set of different encoded bit rates of files served in the workload,

$N_{X_{w_i}}^{\mathrm{memory}}$

is a number of streams in the workload having a memory access to a subset of files encoded at

$X_{W_i}$

Kb/s,

${\mathrm{cost}}_{X_{W_i}}^{\mathrm{memory}}$

is a cost of consumed resources for a stream having a memory access to a file encoded at

$X_{W_i}<br><br>\mathrm{Kb}/s,\text{\hspace{1em}<br><br>}<br><br>N_{X_{w_i}}^{\mathrm{disk}}$

is a number of streams in the workload having a disk access to a subset of files encoded at

$X_{W_i}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{Kb}/s,\mathrm{and}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{cost}}_{X_{w_i}}^{\mathrm{disk}}$

is a cost of consumed resources for a stream having a disk access to a file encoded at

$X_{W_i}$

Kb/s.

29. The method of claim 27 wherein said determining how many servers of said first server configuration to be included at said site for supporting the expected workload in a desired manner comprises:

computing said service demand for resources of a single server of said first server configuration under the expected workload;

determining from said service demand if said single server has sufficient capacity for supporting the expected workload in the desired manner; and

if determined that said single server does not have sufficient capacity for supporting the expected workload in the desired manner, estimating from the service demand a number of servers of said first server configuration that when clustered together provide sufficient capacity for supporting the expected workload in the desired manner.

30. The method of claim 29 further comprising:

computing a second service demand for resources of the cluster of the number of servers under the expected workload.

31. The method of claim 30 further comprising:

analyzing the second service demand to verify whether the estimated number of servers is accurate.

32. The method of claim 24 further comprising:

generating a workload profile for a single one of said first server configuration under the expected workload, wherein said workload profile identifies client accesses in the expected workload that are served from memory of the single one of said first server configuration and client accesses in the expected workload that are served from disk of the first server configuration.

33. The method of claim 32 wherein said determining how many servers of said first server configuration to be included at said site for supporting the expected workload in a desired manner comprises:

using the generated workload profile to evaluate the capacity of said single one of said first server configuration to determine whether said single one of said first server configuration has sufficient capacity for supporting the expected workload in the desired manner.

34. The method of claim 33 further comprising:

if determined that said single one of the said first server configuration does not have sufficient capacity for supporting the expected workload in the desired manner, estimating a number of servers of said first server configuration that when clustered together provide sufficient capacity for supporting the expected workload in the desired manner and generating a second workload profile for the cluster under the expected workload, wherein said second workload profile identifies client accesses in the expected workload that are served from memory of the cluster of servers of said first server configuration and client accesses in the expected workload that are served from disk of the cluster of servers of said first server configuration.

35. The method of claim 34 wherein said generating said second workload profile comprises determining for each of said number of servers in the cluster the client accesses of the expected workload that are assigned to such server under a load balancing strategy.

36. The method of claim 34 further comprising:

using the generated second workload profile to evaluate the capacity of said cluster of servers of said first server configuration to determine whether said cluster has sufficient capacity for supporting the expected workload in the desired manner.

37. A system comprising:

means for receiving workload information representing an expected workload of client accesses of streaming media files from a site; and

means for determining an optimal number of servers of a given configuration type to be implemented at said site for supporting the expected workload.

38. The system of claim 37 wherein said means for determining said optimal number of servers of a given configuration type comprises:

means for determining a workload profile for a server of said given configuration type under the expected workload, wherein said workload profile identifies client accesses in the expected workload that are served from memory of the server of said given configuration type and client accesses in the expected workload that are served from disk of the first server configuration.

39. A method comprising:

receiving workload information identifying an expected workload of client accesses of streaming media files from a server; and

determining a service demand profile for at least one server configuration under evaluation; and

determining from the service demand profile how many of said first server configuration under evaluation are needed for supporting the expected workload in a desired manner.

40. The method of claim 39 wherein said determining a service demand profile for at least one server configuration under evaluation comprises:

computing a cost corresponding to resources of a server that are consumed in supporting the workload, said server having a configuration corresponding to one of said at least one server configuration.

41. The method of claim 40 wherein said computing said cost comprises:

computing a cost of consumed resources for a stream in said workload having a memory access to a streaming media file; and

computing a cost of consumed resources for a stream in said workload having a disk access to a streaming media file.

42. The method of claim 39 wherein said determining a service demand profile for at least one server configuration under evaluation comprises:

determining a service demand profile under the expected workload for resources of a single server having a first configuration corresponding to one of said at least one configuration.

43. The method of claim 42 wherein said determining from the service demand profile how many of said first server configuration under evaluation are needed for supporting the expected workload in a desired manner comprises:

determining from said service demand profile if said single server has sufficient capacity for supporting the expected workload in the desired manner; and

if determined that said single server does not have sufficient capacity for supporting the expected workload in the desired manner, estimating from the service demand a number of servers of said first configuration that when clustered together provide sufficient capacity for supporting the expected workload in the desired manner.

44. The method of claim 43 further comprising:

computing a second service demand profile for resources of the cluster of the number of servers under the expected workload.

45. A system comprising:

a media profiler operable to receive workload information for a service provider'"'"'s site and generate a workload profile for a server configuration under consideration for supporting the service provider'"'"'s site; and

a capacity planner operable to receive the generated workload profile for the server configuration under consideration and determine how many servers of said server configuration are needed to provide sufficient capacity for supporting the site'"'"'s workload.

46. The system of claim 45 wherein in determining how many servers of said server configuration are needed, said capacity planner determines how many servers of said configuration are needed to provide sufficient capacity for supporting the site'"'"'s workload in accordance with at least one service parameter.

47. The system of claim 46 wherein said at least one service parameter comprises information identifying at least one performance criteria desired to be satisfied by the service provider'"'"'s site under the site'"'"'s workload.

48. The system of claim 45 wherein said workload profile comprises:

for a plurality of different points in time, identification of a number of concurrent client accesses, wherein the number of concurrent client accesses are categorized into corresponding encoding bit rates of streaming media files accessed thereby and are further sub-categorized into either memory or disk accesses.

49. The system of claim 45 further comprising:

a dispatcher operable to receive a client access log collected over a period of time for said service provider'"'"'s site and generate said workload information received by said media profiler.

50. The system of claim 49 wherein said dispatcher is operable to receive identification of a number of servers of said server configuration under consideration and determine for each of said number of servers the client accesses of the client access log that are assigned to such server under a load balancing strategy.

51. The system of claim 50 wherein said dispatcher receives input identifying said load balancing strategy.

52. The system of claim 50 wherein said load balancing strategy comprises one selected from the group consisting of:

round-robin and a locality-aware strategy.

53. Computer-executable software code stored to a computer-readable medium, the computer-executable software code comprising:

code for receiving workload information representing an expected workload of client accesses of streaming media files from a site; and

code for determining how many nodes of a given configuration type to be implemented as a cluster at said site for supporting the expected workload in a desired manner.

54. Computer-executable software code of claim 53 wherein said code for determining how many nodes of a given configuration type to be implemented as a cluster at said site for supporting the expected workload in a desired manner comprises:

code for determining how many nodes of said given configuration type is capable of supporting said expected workload in accordance with at least one service parameter.

55. Computer-executable software code of claim 54 wherein said at least one service parameter comprises information identifying at least one performance criteria desired to be satisfied by said site under the expected workload.

56. Computer-executable software code of claim 53 further comprising:

code for generating a workload profile for the received workload information for a node of said given configuration type, wherein said workload profile comprises, for a plurality of different points in time, identification of a number of concurrent client accesses, wherein the number of concurrent client accesses are categorized into corresponding encoding bit rates of streaming media files accessed thereby and are further sub-categorized into either memory or disk accesses.

57. Computer-executable software code of claim 53 wherein said code for how many nodes of a given configuration type to be implemented as a cluster at said site for supporting the expected workload in a desired manner comprises:

code for evaluating a capacity of a plurality of different clusters and determining an optimal one of said plurality of different clusters for supporting the expected workload.