Device and method for parallel simulation task generation and distribution

• US 6,345,240 B1
• Filed: 08/24/1998
• Issued: 02/05/2002
• Est. Priority Date: 08/24/1998
• Status: Expired due to Term

• Current Assignee: Bell Semiconductor, LLC (Hilco Inc. (d/b/a Hilco Global))
• Original Assignee: Agere Systems Guardian Corp. (Broadcom, Inc.)
• Inventors: Havens, Joseph Harold
• Primary Examiner(s): Teska, Kevin J.
• Assistant Examiner(s): Broda, Samuel
• Application Number:
  US09/138,702
• Time in Patent Office: 1,261 Days
• Field of Search: 703/1, 703/6, 703/21, 703/22, 700/100, 700/101, 700/102, 709/100, 709/105
• US Class Current: 703/21
• CPC Class Codes: G06F 2119/06 Power analysis or power optimisation
• CPC Class Codes: G06F 2119/08 Thermal analysis or thermal optimisation
• CPC Class Codes: G06F 2209/503 Resource availability
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals

Abstract

The invention provides a simulation task generator that receives a range of parameters that are desired for a particular parallel simulation. The simulation task generator determines specific combinations of parameters which corresponds to each simulation task and assigns estimated performance ratings for each of the simulation tasks. The simulation task generator retrieves information from a database that indicates availability of accessible processors and ratings corresponding to each of the accessible processors. The simulation task generator matches the simulation tasks to the available processors to achieve an optimum performance. Some of the available processors may also be used as secondary distributors. Such processors may be provided subranges of the parameters and generate simulation tasks and distribute those tasks to other processors more conveniently accessed by the secondary distributor.

Claims

1. A simulation task generator coupled to a network, comprising:

a memory;

a network interface; and

a controller coupled to the memory and the network interface, the controller receiving a parallel simulation request from a user that includes at least one simulation parameter and at least one range of values corresponding to the simulation parameter, the controller generating simulation tasks based on the range of values and distributing the simulation tasks to processor groups to perform independent simulations of a same entity, the processor groups coupled to the network and physically located in different time zones so that working hours corresponding to personnel of the groups are mutually exclusive.

2. The simulation task generator of claim 1, wherein the controller generates one simulation task corresponding to each unique combination of the range of values.

3. The simulation task generator of claim 2, further comprising a database, wherein the controller assigns to the simulation task estimates of processor power and needed resources based on first data either retrieved from the database or input with the parallel simulation request.

4. The simulation task generator of claim 3, wherein the first data retrieved from the database provides information relating to historical processor power and resource requirements for past simulation tasks, the controller generating the estimates for the simulation task based on similarity with the past simulation tasks.

5. The simulation task generator of claim 3, wherein the controller retrieves second data from the database, the second data including information relating to availability of processors accessible via the network for the parallel simulation, the controller allocating the simulation task to one of the processors that is available.

6. The simulation task generator of claim 5, wherein the second data further includes ratings of processor power and resources for each of the accessible processors, the controller allocating the simulation task based on a comparison between the estimates for the simulation task and the ratings of each of the processors that are available for parallel simulation.

7. The simulation task generator of claim 6, wherein the controller identifies one of the processors as a file server for a domain of processors, the controller transmitting simulation task information to the file server for simulation tasks to be executed by all processors of the domain.

8. The simulation task generator of claim 6, wherein the controller identifies at least one of the processors that is available for parallel simulation and that is also capable of acting as a secondary distributor of a first portion of the simulation tasks, the controller assigning a sub-range of the range of values corresponding to the first portion of the simulation tasks to the secondary distributor.

9. The simulation task generator of claim 8, wherein the controller packages a second portion of the simulation tasks that is allocated to a first portion of processors that are to perform the second portion of the simulation tasks and packages the first portion of the simulation tasks that is to be further distributed by the secondary distributor to a second portion of processors.

10. A method for generating simulation tasks to be distributed to processor groups coupled to a network, comprising:

receiving a parallel simulation request from a user that includes at least one simulation parameter and at least one range of values corresponding to the simulation parameter;

generating simulating tasks based on the range of values; and

distributing the simulation tasks to the processor groups to perform independent simulations of a same entity, the processor groups coupled to the network and physically located in different time zones so that working hours corresponding to personnel of the groups are mutually exclusive.

11. The method of claim 10, wherein one simulation task is generated corresponding to each unique combination of the range of values.

12. The method of claim 11, further comprising assigning to the simulation task estimates of processor power and needed resources based on first data either retrieved from a database or input with the parallel simulation request.

13. The method of claim 12, wherein the first data retrieved from the database provides information relating to historical processor power and resource requirements for past simulation tasks, the method further comprising generating the estimates for the simulation task based on similarity with the past simulation tasks.

14. The method of claim 12, further comprising:

retrieving second data from the database, the second data including information relating to availability of processors accessible via the network for the parallel simulation; and

allocating the simulation task to one of the processors that is available.

15. The method of claim 14, further comprising:

identifying one of the processors as a file server for a domain of processors; and

transmitting simulation task information to the file server for simulation tasks to be executed by all processors of the domain.

16. The method of claim 14, wherein the second data further includes ratings of processor power and resources for each of the accessible processors, the method further comprises allocating the simulation task to one of the processors based on a comparison between the estimates for the simulation task and the ratings of each of the processors that are available for parallel simulation.

17. The method of claim 16, further comprising:

identifying at least one of the processors that is available for parallel simulation and that is also capable of acting as a secondary distributor of a first portion of the simulation tasks; and

assigning a sub-range of the range of values corresponding to the first portion of the simulation tasks to the secondary distributor.

18. The method of claim 17, further comprising:

packaging a second portion of the simulation tasks that is allocated to a first portion of processors that are to perform the second portion of the simulation tasks; and

packaging the first portion of the simulation tasks that is to be further distributed by the secondary distributor to a second portion of processors.

19. A simulation task generator coupled to a network, comprising:

a memory;

a network interface; and

a controller coupled to the memory and the network interface, the controller receiving a parallel simulation request from a user that includes at least one simulation parameter and at least one range of values corresponding to the simulation parameter, the controller generating simulation tasks based on the range of values and distributing the simulation tasks to processor groups to perform independent simulations of a same entity, the processor groups coupled to the network and physically located in a single place, the processors of each group being allocated to mutually exclusive time zones.