M
Martin Rinard
Researcher at Massachusetts Institute of Technology
Publications - 381
Citations - 19269
Martin Rinard is an academic researcher from Massachusetts Institute of Technology. The author has contributed to research in topics: Data structure & Compiler. The author has an hindex of 70, co-authored 372 publications receiving 18126 citations. Previous affiliations of Martin Rinard include University of California, Santa Barbara & Stanford University.
Papers
More filters
Using Code Perforation to Improve Performance, Reduce Energy Consumption, and Respond to Failures
TL;DR: The implemented SpeedPress compiler can automatically apply code perforation to existing computations with no developer intervention whatsoever and the result is a transformed computation that can respond almost immediately to a range of increased performance demands while keeping any resulting output distortion within acceptable user-defined bounds.
Proceedings ArticleDOI
Automatic error elimination by horizontal code transfer across multiple applications
TL;DR: Experimental results using seven donor applications to eliminate ten errors in seven recipient applications highlight the ability of CP to transfer code across applications to eliminated out of bounds access, integer overflow, and divide by zero errors.
Proceedings ArticleDOI
Chisel: reliability- and accuracy-aware optimization of approximate computational kernels
TL;DR: Chisel as discussed by the authors is a system for reliability and accuracy-aware optimization of approximate computational kernels that run on approximate hardware platforms, given a combined reliability and/or accuracy specification, automatically selects approximate kernel operations to synthesize an approximate computation that minimizes energy consumption.
Proceedings ArticleDOI
Inference and enforcement of data structure consistency specifications
TL;DR: The results indicate that automatic constraint generation produces constraints that enable programs to execute successfully through data structure consistency errors, and compared to manual specification, automatic generation can produce more comprehensive sets of constraints that cover a larger range ofData structure consistency properties.
Chisel: Reliability- and Accuracy-Aware Optimization of Approximate Computational Kernels
TL;DR: The experimental results show that the implemented optimization algorithm enables Chisel to optimize the authors' set of benchmark kernels to obtain energy savings from 8.7% to 19.8% compared to the fully reliable kernel implementations while preserving important reliability guarantees.