1 / 26

Applying Scheduling and Tuning to On-line Parallel Tomography

Applying Scheduling and Tuning to On-line Parallel Tomography. Shava Smallen Indiana University Henri Casanova, Francine Berman University of California at San Diego San Diego Supercomputer Center. Outline. Introduction to On-line Parallel Tomography Tunable On-line Parallel Tomography

zuzela
Download Presentation

Applying Scheduling and Tuning to On-line Parallel Tomography

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Applying Scheduling and Tuning to On-line Parallel Tomography Shava Smallen Indiana University Henri Casanova, Francine Berman University of California at San Diego San Diego Supercomputer Center

  2. Outline • Introduction to On-line Parallel Tomography • Tunable On-line Parallel Tomography • User-directed application-level scheduler • Experiments • Summary

  3. What is tomography? • Tomography: a method for reconstructing the interior of an object from its projections • National Center for Microscopy and Imaging Research (NCMIR) • Electron Microscopy Electron Microscope

  4. Example • Compute and data-intensive • E.g. 2k x 2k dataset (pixels) • 2k units of work (slices) • Total input data size: 976 MB • Total output data size: 9.6 GB • Compute time: ~ 16 days on a standard workstation • Off-line • Data collection • Data processing • Data viewing Tomogram of spiny dendrite (Images courtesy of Steve Lamont)

  5. On-line Parallel Tomography on-line parallel tomography • Provide interactive soft real-time feedback on quality of data acquisition • High tomogram resolution and frequent refreshes • Efficiency benefits for users and microscope

  6. NCMIR Compute Platform • Distributed multi-user, heterogeneous Grid • Meteor cluster (SDSC) • Pentium III dual procs (Linux) Blue Horizon (SDSC) 1152 procs (AIX, Loadleveler, Maui Scheduler) network • NCMIR cluster • SGI Indigo2, SGI Octane (IRIX) • SUN ULTRA, SUN Enterprise (Solaris)

  7. Application Tunability on-line parallel tomography reduce(f) • On-line parallel tomography is a tunable application • [Chang,et al] Availability of alternate configurations • Resource utilization • Output • On-line parallel tomography output • Tomogram resolution • Refresh frequency • Tunability controlled by configuration pair ( f, r ) where • fis the reduction factor (tomogram resolution) • ris the number of projections per refresh (refresh frequency) • E.g. (2,3)

  8. Tunability/Scheduling • At run-time, we need to find out which configuration pairs are feasible • Flexibility to allow for trade-offs between f and r • e.g., (2, 3 ) or (3, 2) • Resource availability • User bounds • E.g., • Refresh at least once every 10 minutes • Minimum image resolution 256 x 256 pixels • A configuration pair is feasible if we can find a corresponding schedule • We choose an adaptive-scheduling approach

  9. Application-Level Scheduler (AppLeS) • Enable an application to adaptively schedule its execution on distributed, heterogeneous resources in order to improve performance • Type of information used: • static • e.g. application model, network topology, … • dynamic • e.g. Network Weather Service (NWS) - available CPU, bandwidth, … AppLeS + application = self-scheduling application

  10. generate request • User-directed AppLeS • Involves user in scheduling process • Flexible process infeasible adjust request request feasible display pairs review rejects all pairs accepts one find schedule User-directed AppLeS User execute on-line parallel tomography

  11. On-line Parallel Tomography Architecture worker Update tomogram slices worker worker scanlines worker projection worker writer preprocessor

  12. Scheduling Approach • Constrained optimization problem based on soft real-time execution • compute constraint • static benchmark, dynamic CPU availability (NWS) • transfer constraint • topology info (ENV), dynamic bandwidth (NWS) • Problem is a nonlinear program • Exploit small range of fto reduce to multiple mixed integer programs which is solved via lp_solve • approximate solution

  13. Experiments • Goals: • Set 1 – Scheduler Results • Evaluate scheduler efficacy • Evaluate impact of dynamic resource availability on scheduler efficacy • Set 2 – Tunability Results • Evaluate usefulness of tunability • Simulation • Number of experiments • Repeatability

  14. NCMIR Grid • Case Study: • week of traces: May 19 – 26, 2001 • CPU availability (NWS) • Bandwidth (NWS) • Node availability (Maui scheduler showbf)

  15. Scheduling Strategies • 4 scheduling strategies

  16. Simtomo • Simulates an execution of on-line parallel tomography • Uses Simgrid - Casanova [CCGrid’2001] • toolkit for evaluating scheduling algorithms • tasks • resources modeled using traces • E.g. Parameter sweep applications [HCW’00] • 2 types of simulations • Executed at 10 minute intervals • 1004 simulations x 4 schedulers

  17. Simulation Types Real trace 1 1 1 1 1 0 0 0 0 0 1. Partially trace-driven (perfect load predictions) 2 1 3 2. Completely trace-driven (imperfect load predictions) 2. Completely trace-driven (imperfect load predictions) 2. Completely trace-driven (imperfect load predictions) 2 1 3 2 1 3

  18. Performance Metric • Relative refresh lateness expected refresh period (based on r) actual refresh period relative refresh lateness

  19. Scheduling Results (1)(partially trace-driven) May 19-26, 2001 98% Importance of dynamic bandwidth info

  20. Scheduling Results (2)(Completely trace-driven) May 19-26, 2001 57.1% Student Version of MATLAB

  21. Tunability Results • How often does the pair change (i.e., tune) • Assume a single user model where user always chooses pair with lowest f • Find the best pairs throughout simulated week • Snapshot of Monday May 21st • On average, pair changed 25% of the time (2,2) (3,2) (2,2) (3,1) 8:00 9:00 10:00 11:00

  22. Summary • Tunable on-line parallel tomography at NCMIR • Dynamic resource information improves scheduler efficacy • Dynamic bandwidth information is key • Case for tunability in a Grid environment

  23. Future Work • Introduce cost • another tunable parameter: (f, r, $) • More Grid simulations • Traces from various sites across US and Europe • Generalizing to other applications • Rescheduling • Production use at NCMIR

  24. Parallel Tomography at NCMIR projection scanline • Embarrassingly parallel Z specimen slice X Y projection scanline

  25. Scheduling Latency • Time to search for feasible triples 1k x 1k 2k x 2k

More Related