Jatin Chhugani

TL;DR: This work studies a set of three workloads that exemplify the span and complexity of physical simulation applications used in a production environment: fluid dynamics, facial animation, and cloth simulation, which are computationally demanding and can benefit greatly from the acceleration possible with large scale CMPs.

TL;DR: In this article, the authors explore the trade-offs of performing linear algebra using Apache Spark compared to traditional C and MPI implementations on HPC platforms, and apply these methods to 1.6TB particle physics, 2.2TB and 16TB climate modeling and 1.1TB bioimaging data.

TL;DR: This work shows that it can overcome constraints using a combination of efficient low-overhead data compression techniques to reduce transfer volumes along with latency-hiding techniques, resulting in over 6.6X performance improvements over state-of-the-art data transfer techniques, and almost an order of magnitude in energy savings.

TL;DR: In this article, a tree merge sort is performed on the nodes that are cache resident until a block of data migrates to a root node, and the completed output list in memory storage is a list of the fully sorted data.

TL;DR: The GLSC is proposed, which extends scatter-gather hardware to support atomic vector operations and provides an average performance improvement on a set of important RMS kernels of 54% for 4-wide SIMD.