2016
DOI: 10.1007/978-3-319-49748-8_1
|View full text |Cite
|
Sign up to set email alerts
|

Big Data, Simulations and HPC Convergence

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
30
0
2

Year Published

2017
2017
2021
2021

Publication Types

Select...
5
1
1

Relationship

3
4

Authors

Journals

citations
Cited by 40 publications
(32 citation statements)
references
References 11 publications
0
30
0
2
Order By: Relevance
“…To foster faster model convergence, we need to design new collective communication abstractions. We identify all 5 classes of data-intensive computation [2], from pleasingly parallel to machine learning and simulations. To re-design a modular software stack with native kernels to effectively utilize scale-up servers for machine learning and data analytics applications.…”
Section: A Hpc For Machine Learningmentioning
confidence: 99%
See 1 more Smart Citation
“…To foster faster model convergence, we need to design new collective communication abstractions. We identify all 5 classes of data-intensive computation [2], from pleasingly parallel to machine learning and simulations. To re-design a modular software stack with native kernels to effectively utilize scale-up servers for machine learning and data analytics applications.…”
Section: A Hpc For Machine Learningmentioning
confidence: 99%
“…The convergence of HPC and data-intensive methodologies [2] provide a promising approach to major performance improvements. Traditional HPC simulations are reaching the limits of original progress.…”
Section: Introductionmentioning
confidence: 99%
“…The convergence between Big Data and HPC can benefit to businesses, willing to dedicate their HPC environments (or HPC‐like environments, also called enterprise performance computing) to several purposes such as deep learning, artificial intelligence, and Big Data. Researchers and industries already started to exploit this convergence . As an example of a practical case, Kamburugamuve et al exploited HPC high‐performance interconnects, namely, Infiniband and Intel Omni‐Path, in order to improve the performance of the Apache Heron Big Data framework.…”
Section: Introductionmentioning
confidence: 99%
“…HPC workloads are traditionally dominated by parallel applications that are large, parallel and tightly-coupled. Recently, however, other application types have become more prevalent, including dramatically more data-intensive and highthroughput applications and real-time jobs [9], [10]. Therefore, jobs that share the same node can have distinct scheduling objectives and different resource requirements.…”
Section: Introductionmentioning
confidence: 99%