Best Paper Award:

  • Vaivaswatha Nagaraj and R. Govindarajan
    Approximating Flow-Sensitive Pointer Analysis Using Frequent Itemset Mining

Best Student Presentation:

  • Byron Hawkins
    Optimizing Binary Translation for Dynamically Generated Code
  • Long Zheng
    On Performance Debugging of Unnecessary Lock Contentions Multicore Processors: A Replay-based Approach

2015 ACM Student Research Competition Winners:
Graduate Category

    • Gold: William Ogilvie (University of Edinburgh)
      Intelligent Heuristic Construction with Active Learning
    • Silver: Niranjan Hasabnis (Stony Brook University)
      Employing Code Generators as De-code Generators: A Novel Approach for Assembly to IR Translation
    • Bronze: Riyadh Baghdadi (Inria and KU Leuven)
      Extending the Scope of Polyhedral Compilation: Progresses in Handling Irregular Codes and in Scalability

Undergraduate Category

      • Gold: Tharindu Rusira, Milinda Fernando, Chalitha Perera, and Chamara Philips(University of Moratuwa, Sri Lanka)
        Auto-tuning the HotSpot JVM

Test of the Time Award:

SWIFT: Software Implemented Fault Tolerance

The 2005 CGO paper “SWIFT: Software Implemented Fault Tolerance” by George A. Reis, Jonathan Chang, Neil Vachharajani, Ram Rangan, and David I. August marks an important turning point in how program execution can be made more resilient to transient hardware faults. SWIFT enhances the notion of program control-flow integrity to dynamically and flexibly provide software-only, transient-fault tolerance. SWIFT can tune performance and reliability trade-offs to match the needs of the targeted system. In addition, software as a lower-cost and more flexible alternative to traditional hardware techniques has emerged as a necessary solution for building robust systems in the deep nanometer-scale CMOS era. To this end, SWIFT was one of the first attempts to leverage the compiler to improve execution reliability and to improve robustness without any hardware changes or support.