Performance Prediction for Data Intensive Applications on Large Scale Parallel Systems

1. Performance Prediction for Data Intensive Applications on Large Scale Parallel Systems Yuhong Wen and Geoffrey C. Fox Northeast Parallel Architecture Center (NPAC) Syracuse University {wen,gcf}@npac.syr.edu SCI'99 / ISAS'99

2. Why Performance Prediction? • Application complexities • A lot of processors required • Large amount of data involved • Time-consuming processing • Performance prediction to fasten • new parallel computer architecture design • application model design SCI'99 / ISAS'99

3. Performance Prediction Approaches • Concept design level performance prediction • aim to provide a quick and roughly correct performance prediction at the early stage of model design • PetaSIM based on this level • Detailed performance prediction • to provide detailed information of a given application running on specific computer system -- Simulation SCI'99 / ISAS'99

4. PetaSIM Motivation • PetaSIM was designed to allow “qualitative” performance estimates” where in particular the design of machine is particularly easy to change • The project will build up a suite of applications which can be used in future activities such as “Petaflop” architectures studies • Applications are to be derived “by hand” or by automatic generation from Maryland Application Emulators • Special attention to support of hierarchical memory machines and data intensive applications • Support parallelism and representation at different grain sizes • Support simulation of “pure data-parallel” and composition of linked modules SCI'99 / ISAS'99

5. Performance Prediction Process Existing Real Applications Application Emulators PSL Specification Execution Script Architecture Description Cost Model Detailed Simulators PetaSIM Estimator Estimated Performance SCI'99 / ISAS'99

6. Peta-Computing Hierarchy Full Heterogeneous MetaProblem Module Module Module Components Components Module Module Task Parallelism Aggregate Aggregate Data Parallelism Simulate Loosely Synchronize Computation Splitting into Lower Level Memory Hierarchy PetaSIM Real Computing SCI'99 / ISAS'99

7. Performance Prediction Model Application Domain Software / Operating System Domain Hardware Domain Multi Domain model SCI'99 / ISAS'99

8. Three Domain Performance Prediction • Application Domain • to extract the data aggregates • to give abstract data movement and computation behavior • Software / Operating System Domain • to provide the methods for task process and memory management, communication and parallel file access • Hardware Domain • to provide the model of processor and memory components, includes cache as well SCI'99 / ISAS'99

9. Emulators • Extract the application’s computational and data access patterns • A simplified version of the real application, contains all the necessary communication, computation and I/O characteristics • less accurate than full application, but more robust • fast performance prediction for rapid prototyping SCI'99 / ISAS'99

10. PetaSIM Design • We define an object structure for computer (including network) and data • Architecture Description • nodeset & linkset • (describe the architecture memory hierarchy) • Data Description • dataset & distribution • Application Description • execution script • System / Software Description SCI'99 / ISAS'99

11. Architecture Description • A nodeset is a collection of entities with types liked • memory with cache & disks • CPU where results can be calculated • pathway such as bus, switch or network • A linkset connects nodesets together in various ways SCI'99 / ISAS'99

12. Application Description • An application consists of dataset objects • dataset implementation is controled by the distribution objects • application behavior is represented by execution script • a set of command statements • data movements • data computation • synchronization SCI'99 / ISAS'99

13. Nodeset Object Structure • Name: one per nodeset object • type: choose from memory, cache, disk, CPU, pathway • number: number of members of this nodeset in the architecture • grainsize: size in bytes of each member of this nodeset (for memory, cache, disk) • bandwidth: maximum bandwidth allowed in any one member of this nodeset • floatspeed: CPU’s float calculating speed • calculate(): method used by CPU nodeset to perform computation • cacherule: controls persistence of data in a memory or cache • portcount: number of ports on each member of nodeset • portname[]: ports connected to linkset • portlink[]: name of linkset connecting to this port • nodeset_member_list: list of nodeset members in this nodeset (for nodeset member identification) SCI'99 / ISAS'99

14. Linkset Object Structure • Name: one per linkset object • type: choose from updown, across • nodesetbegin: name of initial nodeset joined by this linkset • nodesetend: name of final nodeset joined buy this linkset • topology: used for across networks to specify linkage between members of a single nodeset • duplex: choose from full or half • number: number of members of this linkset in the architecture • latency: time to send zero length message across any member of linkset • bandwidth: maximum bandwidth allowed in any link of this linkset • send(): method that calculates cost of sending a message across the linkset • distribution: name of geometric distribution controlling this linkset • linkset_member_list: list of linkset members in this linkset ( for linkset member identification ) SCI'99 / ISAS'99

15. Dataset Object Structure • Name: one per dataset object • choose from grid1dim, grid2dim, grid3dim, specifies type of dataset • bytesperunit: number of bytes in each unit • floatsperunit: update cost as a floating point arithmetic count • operationsperunit: operations in each unit • update(): method that updates given dataset which is contained in a CPU nodeset and a grainsize controlled by last memory nodeset visited • transmit(): method that calculates cost of transmission of dataset between memory levels either communication or movement up and down hierarchy • Methods can use other parameters or be custom SCI'99 / ISAS'99

16. Execution Script • Currently a few primitives which stress (unlike most languages) movement of data through memory hierarchies • send DATAFAMILY from MEM-LEVEL-L to MEM-LEVEL-K • These reference object names for data and memory nodesets • move DATAFAMILY from MEM-LEVEL-L to MEM-LEVEL-K • Use distribution DISTRIBUTION from MEM-LEVEL-L to MEM-LEVEL-K • compute DATAFAMILY-A, DATAFAMILY-B,… on MEM-LEVEL-L • synchronize (synchronizes all processors --- loosely synchronous barrier) • loop operation SCI'99 / ISAS'99

17. PetaSIM Estimation Schedule • Each nodeset member has its usage control to record when dataset arrives and when to send out to next nodeset member • Each linkset member has its usage control to record at what time the linkset member is free or occupied • Data Driven model: Ready ---> Go (First come, First Service) • Support Both data parallel mode and individual operation on each nodeset, linkset member mode SCI'99 / ISAS'99

18. Architecture of PetaSIM C++ Simulator Multi-User Java Server StandardJava AppletClient StandardJava AppletClient SCI'99 / ISAS'99

19. PetaSIM Experiments IBM-SP2 nodeseet and linkset components -- Architecture I SCI'99 / ISAS'99

20. Titan Estimation Results SCI'99 / ISAS'99

21. Pathfinder Estimation Results SCI'99 / ISAS'99

22. PetaSIM Experiments IBM-SP2 nodeseet and linkset components -- Architecture II SCI'99 / ISAS'99

23. Pathfinder Performance Estimation Results SCI'99 / ISAS'99

24. Pathfinder Estimation Results II SCI'99 / ISAS'99

25. Titan Estimation Results (Fixed) SCI'99 / ISAS'99

26. VMScope Estimation Results SCI'99 / ISAS'99

27. PetaSIM Features • Accurate estimation • Friendly user interface • Easy to modify the architecture design • Easy to monitor the effect of the design change • Fast Estimation • Detail performance estimation • Provide detail usage of each individual nodeset and linkset member in the memory hierarchy SCI'99 / ISAS'99

28. Compare with some other Simulators • Different Simulation Approach • PetaSIM not real run the application, estimate the execution script (operation abstraction) • PetaSIM running on single processor • Similar performance estimation results • PetaSIM can easily deal with different kinds of computer architecture • PetaSIM can get detailed information of any part of the architecture SCI'99 / ISAS'99

29. PetaSIM Current Progress Summary • Architecture Description (nodeset & linkset) • Application Description (dataset & execution script) • Link to Application Emulators • Jacobi hand-written example • Pathfinder, Titan, VMScope real applications (Generated by UMD’s Emulator) • Easy modified Architecture and Application description • Fast and relatively Accurate performance estimation (PetaSIM running on single processor) • Java applet based user Interface • Data Parallel Model & Individual Control SCI'99 / ISAS'99

30. Possible Future Work • Richer set of applications using standard benchmarks and DoD MSTAR • Relate object model to those used in “seamless interfaces” / metacomputing i.e. to efforts to establish (distributed) object model for computation • Review very simple execution script -- should we add more complex primitives or regard “application emulators” as this complex script • Binary format (“compiled PetaSIM”) of architecture and application description ( ASCII format will make execution script very large) • Translation tool from ASCII format to binary format (to retain the friendly user interface) • Upgrade performance evaluation model • Run performance simulation in parallel (i.e. PetaSIM running on multi-processors) SCI'99 / ISAS'99

31. PetaSIM Web-Site URL http://kopernik.npac.syr.edu:4096/petasim/V1.0/PetaSIM.html -- PetaSIM Java Applet front user interface and demo -- Related PetaSIM documents SCI'99 / ISAS'99

32. Interface of PetaSIM Client SCI'99 / ISAS'99