Share This Author
CuSha: vertex-centric graph processing on GPUs
- Farzad Khorasani, Keval Vora, Rajiv Gupta, L. Bhuyan
- Computer ScienceIEEE International Symposium on High-Performance…
- 23 June 2014
CuSha is a CUDA-based graph processing framework that overcomes the above obstacle via use of two novel graph representations: G-Shards and Concatenated Windows.
ABCD: eliminating array bounds checks on demand
- R. Bodík, Rajiv Gupta, Vivek Sarkar
- Computer ScienceACM-SIGPLAN Symposium on Programming Language…
- 1 May 2000
ABCD is a light-weight algorithm for elimination of Array Bounds Checks on Demand that can be applied to a set of frequently executed (hot) bounds checks, which makes it suitable for the dynamic-compilation setting, in which compile-time cost is constrained but hot statements are known.
Frequent value compression in data caches
- Jun Yang, Youtao Zhang, Rajiv Gupta
- Computer ScienceProceedings 33rd Annual IEEE/ACM International…
- 1 December 2000
The design and evaluation of the compression cache (CC) is presented which is a first level cache that has been designed so that each cache line can either hold one uncompressed line or two cache lines which have been compressed to at least half their lengths.
Locating faults through automated predicate switching
- X. Zhang, Neelam Gupta, Rajiv Gupta
- Computer ScienceInternational Conference on Software Engineering
- 28 May 2006
By forcibly switching a predicate's outcome at runtime and altering the control flow, the program state can not only be inexpensively modified, but in addition it is often possible to bring the program execution to a successful completion (i.e., program produces the desired output).
The BALLAST Methodology for Structured Partial Scan Design
An efficient partial scan technique called Ballast (balanced structure scant test) is presented, which leads to a low area overhead and allows 100% coverage of irredundant faults.
Fault localization using value replacement
- D. Jeffrey, Neelam Gupta, Rajiv Gupta
- Computer ScienceInternational Symposium on Software Testing and…
- 20 July 2008
A value profile based approach for ranking program statements according to their likelihood of being faulty, which outperforms Tarantula which is the most effective prior approach for statement ranking based fault localization using the benchmark programs the authors studied.
KickStarter: Fast and Accurate Computations on Streaming Graphs via Trimmed Approximations
- Keval Vora, Rajiv Gupta, G. Xu
- Computer ScienceInternational Conference on Architectural Support…
- 4 April 2017
KickStarter is presented, a runtime technique that can trim the approximate values for a subset of vertices impacted by the deleted edges that works for a class of monotonic graph algorithms and can be readily incorporated in any existing streaming graph system.
Copy or Discard execution model for speculative parallelization on multicores
- Chen Tian, Min Feng, V. Nagarajan, Rajiv Gupta
- Computer Science41st IEEE/ACM International Symposium on…
- 8 November 2008
This paper proposes the Copy or Discard (CorD) execution model in which the state of speculative parallel threads is maintained separately from the nonspeculative computation state, and presents an algorithm for profile-based speculative parallelization that is effective in extracting parallelism from sequential programs.
Precise dynamic slicing algorithms
- X. Zhang, Rajiv Gupta, Youtao Zhang
- Computer Science25th International Conference on Software…
- 3 May 2003
This paper shows that while precise slices can be orders of magnitude smaller than imprecise dynamic slices, for small number of slicing requests, the LP algorithm is faster than an imprecising dynamic slicing algorithm proposed by Agrawal and Horgan.
Cost effective dynamic program slicing
The design and evaluation of a cost effective dynamic program slicing algorithm based upon a dynamic dependence graph representation that is highly compact and rapidly traversable and shows that the dependence graph can be transformed in a manner that increases sharing and sharing can be performed even in the presence of aliasing.