1 / 12

Variation Aware Application Scheduling in Multi-core Systems

Variation Aware Application Scheduling in Multi-core Systems. Lavanya Subramanian, Aman Kumar Carnegie Mellon University {lsubrama, amank}@andrew.cmu.edu Website: http://www.cs.cmu.edu/~amank/. Document Map. Problem Statement Milestones Overview of Project Static Profiling

jenis
Download Presentation

Variation Aware Application Scheduling in Multi-core Systems

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. Variation Aware Application Scheduling in Multi-core Systems Lavanya Subramanian, Aman Kumar Carnegie Mellon University {lsubrama, amank}@andrew.cmu.edu Website: http://www.cs.cmu.edu/~amank/

  2. Document Map • Problem Statement • Milestones • Overview of Project • Static Profiling • Variation Map Construction • Variability incorporation in BLESS • Results

  3. Problem Statement (Aide-mémoire) • The perspective of a chip multi processor being a homogenous set of cores is not a practical one. • A CMP has to be relooked as: • a collection of heterogeneous cores • each core operating at different frequency • each core with a different power profile

  4. Milestones • Milestone 1.1: • Building variability information into the CMP simulator. • Static profiling of applications. • Milestone 2: • Building a scheduler into the CMP simulator. • Milestone 3: • Implementing and analyzing the proposed scheme against the baseline algorithms.

  5. The Lock Stock and Barrel

  6. Static Profiling • Simulate SPEC 2000 benchmarks on Wattch/Sim-GALS • Extract • Memory instruction dynamic power per instruction • Non-memory instruction dynamic power per instruction • Core average leakage power per cycle Power Model

  7. Static profiling (Results) Tech: 45 nm Sim GALS

  8. Variation Map Construction • Generate Leff variation map from Varimap tool • Calculate Leakage Variation • Based on Leff variation using SPICE and MATLAB • Calculate Frequency Variation (Base : 3GHz) • Based on Leff variation using MATLAB Tech: 45 nm Frequency Model (Built-in Variations) Power Model (Built-in Variations)

  9. Variability per CORE • Read the Frequency/Leakage maps in BLESS • Compute Power/Performance based on Variability information Frequency Model (Built-in Variations) C1 C2 C3 C4 C1 C2 C3 C4 C5 C6 C7 C8 C5 C6 C7 C8 Power Model (Built-in Variations) C9 C10 C11 C12 C9 C10 C11 C12 C13 C14 C15 C16 C13 C14 C15 C16

  10. Variability per CORE (Results) • Same Application on 16 Processors • 4 Applications on 16 Processors

  11. The Lock Stock and Barrel

  12. Questions

More Related