Results 51 to 60 of about 6,589 (215)

Rcmp: Reconstructing RDMA-Based Memory Disaggregation via CXL

ACM Transactions on Architecture and Code Optimization (TACO)
Memory disaggregation is a promising architecture for modern datacenters that separates compute and memory resources into independent pools connected by ultra-fast networks, which can improve memory utilization, reduce cost, and enable elastic scaling of
Zhonghua Wang, Yixing Guo, Kai Lu, Jiguang Wan, Daohui Wang, Ting Yao, Huatao Wu   +6 more
semanticscholar   +1 more source

APEnet+: high bandwidth 3D torus direct network for petaflops scale commodity clusters

, 2011
We describe herein the APElink+ board, a PCIe interconnect adapter featuring the latest advances in wire speed and interface technology plus hardware support for a RDMA programming model and experimental acceleration of GPU networking; this design allows
, , A Biagioni, A Lonardo, A Salamon, Ammendola R, Ammendola R, Ammendola R, Ammendola R, Ammendola R, Bodin F, Chalasani Suresh, D Rossetti, F Lo Cicero, F Simula, G Salina, L Tosoratto, NVIDIA Corporation, O Prezza, P S Paolucci, P Vicini, Paolucci P S, Paolucci P S, R Ammendola   +23 more
core   +1 more source

Evaluating the benefits of key-value databases for scientific applications [PDF]

, 2019
The convergence of Big Data applications with High-Performance Computing requires new methodologies to store, manage and process large amounts of information. Traditional storage solutions are unable to scale and that results in complex coding strategies.
Becerra Fontal, Yolanda, Gil, Eloy, Glock, Philipp, Oden, Lena, Santamaria Mateu, Pol, Sirvent, Raül, Torres Viñals, Jordi   +6 more
core   +1 more source

Traditional vs. Modern Data Paths: A Comprehensive Survey

Computers, 2022
Recently, many new network paths have been introduced while old paths are still in use. The trade-offs remain vague and should be further addressed. Since last decade, the Internet is playing a major role in people’s lives, and the demand on the Internet
Ahmad Barghash, Lina Hammad, Ammar Gharaibeh   +2 more
doaj   +1 more source

Feasibility tests of RoCE v2 for LHCb event building [PDF]

EPJ Web of Conferences, 2020
This paper evaluates the utilization of Remote Direct Memory Access (RDMA) over Converged Ethernet (RoCE) for the Run 3 LHCb event building at CERN.
Krawczyk Rafał Dominik, Colombo Tommaso, Neufeld Niko, Pisani Flavio, Valat Sébastien   +4 more
doaj   +1 more source

RDMA vs. RPC for implementing distributed data structures [PDF]

, 2019
Distributed data structures are key to implementing scalable applications for scientific simulations and data analysis. In this paper we look at two implementation styles for distributed data structures: remote direct memory access (RDMA) and remote ...
Brock, BA, Buluç, A, Chen, Y, Owens, J, Yan, J, Yelick, K   +5 more
core   +3 more sources

TensorFlow Doing HPC

, 2019
TensorFlow is a popular emerging open-source programming framework supporting the execution of distributed applications on heterogeneous hardware. While TensorFlow has been initially designed for developing Machine Learning (ML) applications, in fact ...
Bulatov, Yaroslav, Chien, Steven W. D., Laure, Erwin, Markidis, Stefano, Olshevsky, Vyacheslav, Vetter, Jeffrey S.   +5 more
core   +1 more source

Light‐activated 3D printed fish‐like actuator

Journal of Applied Polymer Science, Volume 141, Issue 32, August 20, 2024.
3D printing was used to create hydrogel‐based actuator element with gelatin methacryloyl and poly(N‐isopropylacrylamide) in layered rectangles. This unit was integrated in the tail of a fish‐shaped actuator with gold nanostars as photothermal components.
Stefan Pendlmayr, Miguel Alexandre Ramos‐Docampo, David H. Bento, Edit Brodszkij, Dennis Wilkens Juhl, Niels Chr Nielsen, Brigitte Städler   +6 more
wiley   +1 more source

Survey on Traffic Management in Data Center Network: From Link Layer to Application Layer

IEEE Access, 2021
Due to the explosive growth of all kinds of Internet services, data centers have become an irreplaceable and vital infrastructure to support this soaring trend. Compared with traditional networks, data center networks (DCNs) have unique features, such as
Weihe Li, Jingling Liu, Shiqi Wang, Tao Zhang, Shaojun Zou, Jinbin Hu, Wanchun Jiang, Jiawei Huang   +7 more
doaj   +1 more source

Scalable Distributed DNN Training using TensorFlow and CUDA-Aware MPI: Characterization, Designs, and Performance Evaluation

, 2018
TensorFlow has been the most widely adopted Machine/Deep Learning framework. However, little exists in the literature that provides a thorough understanding of the capabilities which TensorFlow offers for the distributed training of large ML/DL models ...
Awan, Ammar Ahmad, Bedorf, Jeroen, Chu, Ching-Hsiang, Panda, Dhabaleswar K., Subramoni, Hari   +4 more
core   +1 more source