Výsledky vyhledávání - "Mohror, Kathryn"

Report

Formal Definitions and Performance Comparison of Consistency Models for Parallel File Systems

Autor: Wang, Chen, Mohror, Kathryn, Snir, Marc

Publikováno v: IEEE Transactions on Parallel and Distributed Systems, 2024, Volume 35, Issue 6, Pages 937-951

The semantics of HPC storage systems are defined by the consistency models to which they abide. Storage consistency models have been less studied than their counterparts in memory systems, with the exception of the POSIX standard and its strict consi

Externí odkaz: http://arxiv.org/abs/2402.14105

Zobrazit plný text záznamu

Report

ML-based Modeling to Predict I/O Performance on Different Storage Sub-systems

Autor: Xu, Yiheng, Sivaraman, Pranav, Devarajan, Hariharan, Mohror, Kathryn, Bhatele, Abhinav

Parallel applications can spend a significant amount of time performing I/O on large-scale supercomputers. Fast near-compute storage accelerators called burst buffers can reduce the time a processor spends performing I/O and mitigate I/O bottlenecks.

Externí odkaz: http://arxiv.org/abs/2312.06131

Zobrazit plný text záznamu

Akademický článek

I/O Aware Power Shifting

Autor: Savoie, Lee, Lowenthal, David K., Supinski, Bronis R. de, Islam, Tanzima, Mohror, Kathryn, Rountree, Barry, Schulz, Martin

Power limits on future high-performance computing (HPC) systems will constrain applications. However, HPC applications do not consume constant power over their lifetimes. Thus, applications assigned a fixed power bound may be forced to slow down duri

Externí odkaz: http://hdl.handle.net/10150/622666
http://arizona.openrepository.com/arizona/handle/10150/622666

Zobrazit plný text záznamu

Report

VELOC: VEry Low Overhead Checkpointing in the Age of Exascale

Autor: Nicolae, Bogdan, Moody, Adam, Kosinovsky, Gregory, Mohror, Kathryn, Cappello, Franck

Publikováno v: SuperCheck'21: First International Symposium on Checkpointing for Supercomputing, 2021

Checkpointing large amounts of related data concurrently to stable storage is a common I/O pattern of many HPC applications. However, such a pattern frequently leads to I/O bottlenecks that lead to poor scalability and performance. As modern HPC infr

Externí odkaz: http://arxiv.org/abs/2103.02131

Zobrazit plný text záznamu