System and method for parallel support of multidimensional slices with a multidimensional database
A system and method is described for use with a multidimensional database computing environment to provide support for parallel calculation of multidimensional slices. Users are able to specify a set of slices and a number of parallel threads to employ. The multidimensional database environment generates tasks that include calculations and/or aggregations, which are able to be executed by the system in parallel. Also described herein are mechanisms of utilizing variables within the calculations performed by respective threads, and combining thread variables after execution.
1. A system for providing parallel calculation within a multidimensional database environment, comprising:
a multidimensional database for storing data associated with a plurality of dimensions, each dimension having a plurality of members; and
a computing device having a processor executing instructions that configures the processor to:
receive a calculation script specifying at least a member list and a command to be performed on a portion of the multidimensional database indicated by the member list,
determine two or more slices to be calculated based upon at least the member list,
generate a task list for the calculation script based on the determined two or more slices that contains a plurality of tasks that are simultaneously executable, wherein each task of the plurality of tasks includes the command and is associated with a respective slice of the two or more slices, and
queue each task of the task list in an order to be executed by a set of threads at a thread pool.
2. The system of claim 1 ,
wherein each task is queued according to an order indicated by the received member list; and
wherein the each task is removed from the queue by a thread of the thread pool in order of the queue.
3. The system of claim 2 ,
wherein each of the two or more slices comprises a unique combination of members from the member list.
4. The system of claim 1 ,
wherein two or more of the tasks on the task list are executed by threads of the thread pool in parallel.
5. The system of claim 1 ,
wherein each task is queued according to a database outline order; and
wherein the each task is removed from the queue by a thread of the thread pool in order of the queue.
6. The system of claim 5 ,
wherein each of the two or more slices comprises a unique combination of members from the member list; and
wherein two or more of the tasks on the task list are executed by threads of the thread pool in parallel.
7. A method for parallel calculation for a multidimensional database, comprising:
providing a multidimensional database for storing data associated with a plurality of dimensions, each dimension having a plurality of members; and
receiving a calculation script specifying at least a member list and a command to be performed on a portion of the multidimensional database indicated by the member list;
determining two or more slices to be calculated based upon at least the member list;
generating a task list based on the determined two or more slices that contains a plurality of tasks that are simultaneously executable, wherein each task of the plurality of tasks includes the command specified in the calculation script and is associated with a respective slice of the two or more slices; and
queuing each task of the task list in an order to be executed by a set of threads at a thread pool.
8. The method of claim 7 ,
wherein each task is queued according to an order indicated by the received member list; and
wherein the each task is removed from the queue by a thread of the thread pool in order of the queue.
9. The method of claim 8 ,
wherein each of the two or more slices comprises a unique combination of members from the member list.
10. The method of claim 7 ,
wherein two or more of the tasks on the task list are executed by threads of the thread pool in parallel.
11. The method of claim 7 ,
wherein each task is queued according to a database outline order; and
wherein the each task is removed from the queue by a thread of the thread pool in order of the queue.
12. The method of claim 11 ,
wherein each of the two or more slices comprises a unique combination of members from the member list; and
wherein two or more of the tasks on the task list are executed by threads of the thread pool in parallel.
13. A non-transitory computer-readable storage medium having instructions stored thereon for parallel calculation for a multidimensional database, which when read and executed by a computer, cause the computer perform steps comprising:
providing a multidimensional database for storing data associated with a plurality of dimensions, each dimension having a plurality of members; and
receiving a calculation script specifying at least a member list and a command to be performed on a portion of the multidimensional database indicated by the member list;
determining two or more slices to be calculated based upon at least the member list;
generating a task list for the calculation script based on the determined two or more slices that contains a plurality of tasks that are simultaneously executable, wherein each task of the plurality of tasks includes the command and is associated with a respective slice of the two or more slices; and
queuing each task of the task list in an order to be executed by a set of threads at a thread pool.
14. The non-transitory computer-readable storage medium of claim 13 ,
wherein each task is queued according to an order indicated by the received member list; and
wherein the each task is removed from the queue by a thread of the thread pool in order of the queue.
15. The non-transitory computer-readable storage medium of claim 14 ,
wherein each of the two or more slices comprises a unique combination of members from the member list.
16. The non-transitory computer-readable storage medium of claim 13 ,
wherein two or more of the tasks on the task list are executed by threads of the thread pool in parallel.
17. The non-transitory computer-readable storage medium of claim 13 ,
wherein each task is queued according to a database outline order; and
wherein the each task is removed from the queue by a thread of the thread pool in order of the queue; and
wherein each of the two or more slices comprises a unique combination of members from the member list.