P
Paul Hargrove
Researcher at Lawrence Berkeley National Laboratory
Publications - 27
Citations - 950
Paul Hargrove is an academic researcher from Lawrence Berkeley National Laboratory. The author has contributed to research in topics: Partitioned global address space & Programming paradigm. The author has an hindex of 14, co-authored 25 publications receiving 864 citations.
Papers
More filters
Journal ArticleDOI
Berkeley Lab Checkpoint/Restart (BLCR) for Linux Clusters
Paul Hargrove,Jason Duell +1 more
TL;DR: The motivation, design and implementation of Berkeley Lab Checkpoint/Restart (BLCR), a system-level checkpoint/restart implementation for Linux clusters that targets the space of typical High Performance Computing applications, including MPI, are described.
Proceedings ArticleDOI
CIFTS: A Coordinated Infrastructure for Fault-Tolerant Systems
Rinku Gupta,Pete Beckman,Byung-Hoon Park,Ewing Lusk,Paul Hargrove,Al Geist,Dhabaleswar K. Panda,Andrew Lumsdaine,Jack Dongarra +8 more
TL;DR: This paper proposes a coordinated infrastructure, named CIFTS, that enables system software components to share fault information with each other and adapt to faults in a holistic manner and demonstrates the nonintrusive low-overhead capability of CIF TS that lets applications run with minimal performance degradation.
Proceedings ArticleDOI
Scaling communication-intensive applications on BlueGene/P using one-sided communication and overlap
TL;DR: It is demonstrated that the PGAS model, using a new port of the Berkeley UPC compiler and GASNet one-sided communication layer, outperforms two-sided (MPI) communication in both microbenchmarks and a case study of the communication-limited benchmark, NAS FT.
Proceedings ArticleDOI
Efficient data race detection for distributed memory parallel programs
TL;DR: The framework for the UPC programming language is implemented and scalability up to a thousand cores for programs with both fine-grained and bulk (MPI style) communication and single- and split-phase barriers is demonstrated.
Proceedings ArticleDOI
Hybrid PGAS runtime support for multicore nodes
TL;DR: This paper presents a new process-based shared memory runtime for PGAS languages, and shows speedups of over 60% for application benchmarks and 100% for collective communication benchmarks, when compared to the previous pthreads implementation.