English
Related papers

Related papers: pMR: A high-performance communication library

200 papers

Since local LLM inference on resource-constrained edge devices imposes a severe performance bottleneck, this paper proposes distributed prompt caching to enhance inference performance by cooperatively sharing intermediate processing states…

Machine Learning · Computer Science 2026-04-13 Hiroki Matsutani , Naoki Matsuda , Naoto Sugiura

We evaluate optimized parallel sparse matrix-vector operations for two representative application areas on widespread multicore-based cluster configurations. First the single-socket baseline performance is analyzed and modeled with respect…

Performance · Computer Science 2012-03-01 Gerald Schubert , Georg Hager , Holger Fehske , Gerhard Wellein

Isomorphic (sparse) collective communication is a form of collective communication in which all involved processes communicate in small, identically structured neighborhoods of other processes. Isomorphic neighborhoods are defined via an…

Distributed, Parallel, and Cluster Computing · Computer Science 2016-06-27 Jesper Larsson Träff , Alexandra Carpen-Amarie , Sascha Hunold , Antoine Rougier

Apache Spark is a popular system aimed at the analysis of large data sets, but recent studies have shown that certain computations---in particular, many linear algebra computations that are the basis for solving common machine learning…

Distributed, Parallel, and Cluster Computing · Computer Science 2018-05-31 Alex Gittens , Kai Rothauge , Shusen Wang , Michael W. Mahoney , Lisa Gerhardt , Prabhat , Jey Kottalam , Michael Ringenburg , Kristyn Maschhoff

Processing-in-memory (PIM) architectures have demonstrated great potential in accelerating numerous deep learning tasks. Particularly, resistive random-access memory (RRAM) devices provide a promising hardware substrate to build PIM…

Hardware Architecture · Computer Science 2022-02-01 Weidong Cao , Yilong Zhao , Adith Boloor , Yinhe Han , Xuan Zhang , Li Jiang

Asynchronous Many-Task (AMT) runtimes offer a productive alternative to the Message Passing Interface (MPI). However, the diverse AMT landscape makes fair comparisons challenging. Task Bench, proposed by Slaughter et al., addresses this…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-03-16 Torben R. Lahnor , Mia Reitz , Jonas Posner , Patrick Diehl

Many ML applications and products train on medium amounts of input data but get bottlenecked in real-time inference. When implementing ML systems, conventional wisdom favors segregating ML code into services queried by product code via…

Machine Learning · Computer Science 2023-07-25 Daniel S Johnson , Igor L Markov

We present MT-lib, an efficient message transfer library for messages gather and scatter in benchmarks like Graph500 for Supercomputers. Our library includes MST version as well as new-MST version. The MT-lib is deliberately kept…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-03-30 Xinbiao Gan , Wen Tan

With the wide adoption of language models for IR -- and specifically RAG systems -- the latency of the underlying LLM becomes a crucial bottleneck, since the long contexts of retrieved passages lead large prompts and therefore, compute…

Information Retrieval · Computer Science 2026-04-06 Cornelius Kummer , Lena Jurkschat , Michael Färber , Sahar Vahdati

The true costs of high performance computing are currently dominated by software. Addressing these costs requires shifting to high productivity languages such as Matlab. MatlabMPI is a Matlab implementation of the Message Passing Interface…

Astrophysics · Physics 2007-05-23 Jeremy Kepner , Stan Ahalt

We present an easy to use and flexible grid library for developing highly scalable parallel simulations. The distributed cartesian cell-refinable grid (dccrg) supports adaptive mesh refinement and allows an arbitrary C++ class to be used as…

Distributed, Parallel, and Cluster Computing · Computer Science 2013-01-30 I. Honkonen , S. von Alfthan , A. Sandroos , P. Janhunen , M. Palmroth

A high performance Beowulf (PC cluster) machine installed with Linux operating system and MPI (Message Passing Interface) for interprocessor communications has been constructed using Gigabit Ethernet and the communication software GAMMA…

Computational Physics · Physics 2007-05-23 Motohiko Tanaka

Traditional parallel schedulers running on cluster supercomputers support only static scheduling, where the number of processors allocated to an application remains fixed throughout the execution of the job. This results in…

Distributed, Parallel, and Cluster Computing · Computer Science 2007-06-15 Rajesh Sudarsan , Calvin J. Ribbens

We present Matrix Distributed Processing, a C++ library for fast development of efficient parallel algorithms. MDP is based on MPI and consists of a collection of C++ classes and functions such as lattice, site and field. Once an algorithm…

Distributed, Parallel, and Cluster Computing · Computer Science 2007-05-23 Massimo Di Pierro

Large model inference is shifting from cloud to edge due to concerns about the privacy of user interaction data. However, edge devices often struggle with limited computing power, memory, and bandwidth, requiring collaboration across…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-10-02 Zonghang Li , Wenjiao Feng , Mohsen Guizani , Hongfang Yu

With the recent success of large language models (LLMs), the idea of AI-augmented Business Process Management systems is becoming more feasible. One of their essential characteristics is the ability to be conversationally actionable,…

Artificial Intelligence · Computer Science 2026-04-21 Nataliia Klievtsova , Timotheus Kampik , Juergen Mangler , Stefanie Rinderle-Ma

The critical path of internode communication on large-scale systems is composed of multiple components. When a supercomputing application initiates the transfer of a message using a high-level communication routine such as an MPI_Send, the…

Distributed, Parallel, and Cluster Computing · Computer Science 2020-02-10 Rohit Zambre , Megan Grodowitz , Aparna Chandramowlishwaran , Pavel Shamis

Decreasing sequence length is a common way to accelerate transformers, but prior token reduction work often targets classification and reports proxy metrics rather than end-to-end latency. For semantic segmentation, token reduction is…

Computer Vision and Pattern Recognition · Computer Science 2026-04-08 Simon Ravé , Pejman Rasti , David Rousseau

The collective operations are considered critical for improving the performance of exascale-ready and high-performance computing applications. On this paper we focus on the Message-Passing Interface (MPI) Allgather many to many collective,…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-09-21 Wilton Jaciel Loch , Guilherme Piêgas Koslovski

Reservoir computing systems rely on the recurrent multiplication of a very large, sparse, fixed matrix. We argue that direct spatial implementation of these fixed matrices minimizes the work performed in the computation, and allows for…

Hardware Architecture · Computer Science 2021-01-25 Matthew Denton , Herman Schmit