Zobrazeno 1 - 7
of 7
pro vyhledávání: '"Saurav Muralidharan"'
Publikováno v:
IEEE Micro. 40:17-25
Deep neural networks (DNNs) frequently contain far more weights, represented at a higher precision, than are required for the specific task which they are trained to perform. Consequently, they can often be compressed using techniques such as weight
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::33ff3ddc5bbf827319388fd29e93ed9b
https://doi.org/10.1109/mm.2020.3012391
https://doi.org/10.1109/mm.2020.3012391
Publikováno v:
ACM Transactions on Architecture and Code Optimization. 13:1-24
This article describes Surge, a nested data-parallel programming system designed to simplify the porting and tuning of parallel applications to multiple target architectures. Surge decouples high-level specification of computations, expressed using a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::004d5f7e5e7d43ab112d9cd761a9a8b3
https://doi.org/10.1145/3012011
https://doi.org/10.1145/3012011
Publikováno v:
ASPLOS
Code variants represent alternative implementations of a computation, and are common in high-performance libraries and applications to facilitate selecting the most appropriate implementation for a specific execution context (target architecture and
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6d63288cf164f7c65de34878623b802d
https://doi.org/10.1145/2980024.2872411
https://doi.org/10.1145/2980024.2872411
Autor:
Saurav Muralidharan, Protonu Basu, Manu Shantharam, Shreyas Ramalingam, Anand Venkat, Suchit Maindola, Axel Rivera, Malik Muhammad Zaki Murtaza Khan, Mary Hall
Publikováno v:
The International Journal of High Performance Computing Applications. 27:379-393
Autotuning systems employ empirical techniques to evaluate the suitability of a search space of possible implementations of a computation. Autotuning has emerged as a critical strategy for achieving high performance as architectural complexity grows.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::739032c28ff756a36f6df6e92c9c82e0
https://doi.org/10.1177/1094342013493644
https://doi.org/10.1177/1094342013493644
Publikováno v:
PPOPP
This paper describes Surge, a collection-oriented programming model that enables programmers to compose parallel computations using nested high-level data collections and operators. Surge exposes a code generation interface, decoupled from the core c
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::bbc855eccee4f9af7ac0f57f50b8c6fb
https://doi.org/10.1145/2688500.2688537
https://doi.org/10.1145/2688500.2688537
Publikováno v:
IPDPS
Auto tuning systems intelligently navigate a search space of possible implementations of a computation to find the implementation(s) that best meets a specific optimization criteria, usually performance. This paper describes Nitro, a programmer-direc
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::933423a730fadf14099ce21f475ef105
https://doi.org/10.1109/ipdps.2014.59
https://doi.org/10.1109/ipdps.2014.59
Publikováno v:
NetCoM
Private P2P networks are peer-to-peer (P2P) networks that only allow some mutually trusted computers to share resources. In this paper, we consider a special type of Private P2P system - the Secure-group P2P (SGP) system. SGP systems form an overlay
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::da8fbd82583dbac79d54248ec201198d
https://doi.org/10.1109/netcom.2009.41
https://doi.org/10.1109/netcom.2009.41