System and method for evaluating a capacity of a streaming media server for supporting a workload
AssigneeHewlett-Packard Development Company, L.P.
InventorsLudmila Cherkasova, Wenting Tang
Life of this patent
Estimating the term…
Abstract
A method comprises receiving, into a capacity planning tool, configuration information for at least one streaming media server. The method further comprises receiving, into the capacity planning tool, workload information for a workload of client accesses of streaming media files from a server, and the capacity planning tool evaluating a capacity of the at least one streaming media server for supporting the workload.
Claims30
A method comprising:
- receiving, into a capacity planning tool executed in a computer system, configuration information for at least one streaming media server, wherein the configuration information comprises a single file benchmark and a unique file benchmark for the at least one streaming media server, wherein the single file benchmark measures a streaming media server capacity when all clients in a test workload are accessing a single file, and wherein the unique file benchmark measures the streaming media server capacity when each client in the test workload is accessing a different file;
- receiving, into said capacity planning tool, workload information for a workload of client accesses of streaming media files from a server;
- and
- said capacity planning tool evaluating, based on said configuration information, a capacity of the at least one streaming media server for supporting the workload.
A method comprising:
- receiving, into a capacity planning tool executed in a computer system, configuration information for at least one streaming media server;
- receiving, into said capacity planning tool, workload information for a workload of client accesses of streaming media files from a server;
- said capacity planning tool evaluating a capacity of the at least one streaming media server for supporting the workload;
- wherein said evaluating comprises computing a service demand for said at least one streaming media server supporting said workload;
- and
- wherein said computing said service demand comprises computing:
- Demand
- =
- ∑
- i
- =
- 1
- K
- W
-
- N
- X
- W
- i
- memory
- ×
- cost
- X
- W
- i
- memory
- +
- ∑
- i
- =
- 1
- K
- W
-
- N
- X
- W
- i
- disk
- ×
- cost
- X
- W
- i
- disk
- ,
- wherein the workload W comprises XW=X1, . . . , Xkw set of different encoded bit rates of files served in the workload,
- N
- X
- W
- i
- memory
- is a number of streams in the workload having a memory access to a subset of files encoded at XWi Kb/s,
- cos
-
-
- t
- X
- W
- i
- memory
- is a cost of consumed resources for a stream having a memory access to a file encoded at XWi Kb/s,
- N
- X
- W
- i
- memory
- is a number of streams in the workload having a disk access to a subset of files encoded at XWi Kb/s, and
- cost
- X
- W
- i
- memory
- is a cost of consumed resources for a stream having a disk access to a file encoded at XWi Kb/s.
Computer-executable software code stored to a computer-readable medium, the computer-executable software code comprising:
- code executable in a computer system for receiving workload information for a workload of client accesses of streaming media files from a server;
- and
- code executable in the computer system for employing a cost function derived for at least one system configuration from a single file benchmark and a unique file benchmark for evaluating a capacity of the at least one system configuration for supporting the workload, wherein the single file benchmark measures a streaming media server capacity when all clients in a test workload are accessing a single file, and wherein the unique file benchmark measures the streaming media server capacity when each client in the test workload is accessing a different file.
A system comprising:
- means for receiving configuration information for a plurality of different system configurations, wherein the configuration information comprises, for each of the plurality of different system configurations, a corresponding single file benchmark and unique file benchmark, wherein said single file benchmark measures capacity of the corresponding system configuration for serving a population of clients that all access a same file, wherein said unique file benchmark measures capacity of the corresponding system configuration for serving a population of clients that all access different files;
- means for receiving workload information for a workload of client accesses of streaming media files from a server;
- and
- means for evaluating, based on the configuration information, the capacity of each of said plurality of different system configurations for supporting said workload.
A method executable by a computer system, comprising:
- determining results of a single file benchmark for each of a plurality of encoding bit rates of a single file served by at least a first streaming media server configuration, wherein the result of the single file benchmark for a given encoding bit rate identifies the maximum number of concurrent streams of the single file that the at least a first streaming media server configuration can supply to a population of clients at the given encoding bit rate;
- determining results of a unique file benchmark for each of said plurality of encoding bit rates, wherein the result of the unique file benchmark for a given encoding bit rate identifies the maximum number of concurrent streams of different files that the at least a first streaming media server configuration can supply to the population of clients at the given encoding bit rate;
- deriving, from the results of the single file benchmark and unique file benchmark, a cost function;
- receiving, into a capacity planning tool, workload information for a workload of client accesses of streaming media files from a server;
- and
- using, by the capacity planning tool, the cost function for said at least a first streaming media server configuration for evaluating a capacity of the at least a first streaming media server configuration for supporting the workload.
Dependencies are inferred from the claim text.