Zobrazeno 1 - 10
of 10
pro vyhledávání: '"Rakesh Komuravelli"'
Autor:
Mark Zhao, Niket Agarwal, Aarti Basant, Buğra Gedik, Satadru Pan, Mustafa Ozdal, Rakesh Komuravelli, Jerry Pan, Tianshu Bao, Haowei Lu, Sundaram Narayanan, Jack Langman, Kevin Wilfong, Harsha Rastogi, Carole-Jean Wu, Christos Kozyrakis, Parik Pol
Publikováno v:
Proceedings of the 49th Annual International Symposium on Computer Architecture.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::cad44daa1e8b7357b692548a25592146
https://doi.org/10.1145/3470496.3533044
https://doi.org/10.1145/3470496.3533044
Autor:
Dheevatsa Mudigere, Yuchen Hao, Jianyu Huang, Zhihao Jia, Andrew Tulloch, Srinivas Sridharan, Xing Liu, Mustafa Ozdal, Jade Nie, Jongsoo Park, Liang Luo, Jie (Amy) Yang, Leon Gao, Dmytro Ivchenko, Aarti Basant, Yuxi Hu, Jiyan Yang, Ehsan K. Ardestani, Xiaodong Wang, Rakesh Komuravelli, Ching-Hsiang Chu, Serhat Yilmaz, Huayu Li, Jiyuan Qian, Zhuobo Feng, Yinbin Ma, Junjie Yang, Ellie Wen, Hong Li, Lin Yang, Chonglin Sun, Whitney Zhao, Dimitry Melts, Krishna Dhulipala, KR Kishore, Tyler Graf, Assaf Eisenman, Kiran Kumar Matam, Adi Gangidi, Guoqiang Jerry Chen, Manoj Krishnan, Avinash Nayak, Krishnakumar Nair, Bharath Muthiah, Mahmoud khorashadi, Pallab Bhattacharya, Petr Lapukhov, Maxim Naumov, Ajit Mathews, Lin Qiao, Mikhail Smelyanskiy, Bill Jia, Vijay Rao
Deep learning recommendation models (DLRMs) are used across many business-critical services at Facebook and are the single largest AI application in terms of infrastructure demand in its data-centers. In this paper we discuss the SW/HW co-designed so
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::3b1a6d07c5739fe931da1f91ac4f756a
http://arxiv.org/abs/2104.05158
http://arxiv.org/abs/2104.05158
Autor:
Sarita V. Adve, Vikram Adve, Matthew D. Sinclair, Prakalp Srivastava, Maria Kotsifakou, Rakesh Komuravelli
Publikováno v:
PPOPP
We propose a parallel program representation for heterogeneous systems, designed to enable performance portability across a wide range of popular parallel hardware, including GPUs, vector instruction sets, multicore CPUs and potentially FPGAs. Our re
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::bbd9f342fb2a1449b13fa7d5472f9cbb
https://doi.org/10.1145/3200691.3178493
https://doi.org/10.1145/3200691.3178493
Autor:
Vikram Adve, Johnathan Alsop, Rakesh Komuravelli, Maria Kotsifakou, Matthew D. Sinclair, Prakalp Srivastava, Muhammad Huzaifa, Sarita V. Adve
Publikováno v:
ISCA
Heterogeneous systems employ specialization for energy efficiency. Since data movement is expected to be a dominant consumer of energy, these systems employ specialized memories (e.g., scratchpads and FIFOs) for better efficiency for targeted data. T
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::52a43b10002430d8d5729412959bf62e
https://doi.org/10.1145/2872887.2750374
https://doi.org/10.1145/2872887.2750374
Publikováno v:
IEEE Micro. 34:138-148
Recent research in disciplined shared-memory programming models presents a unique opportunity for rethinking the multicore memory hierarchy for better efficiency in terms of complexity, performance, and energy. The DeNovo hardware system showed that
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::affd680f5e8e8c42c54ffa356847e87e
https://doi.org/10.1109/mm.2014.5
https://doi.org/10.1109/mm.2014.5
Publikováno v:
ISPASS
In recent years the power wall has prevented the continued scaling of single core performance. This has lead to the rise of dark silicon and motivated a move toward parallelism and specialization. As a result, energy-efficient high-throughput GPU cor
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::6d67f0ba03d00811ed043cf216218652
https://doi.org/10.1109/ispass.2016.7482092
https://doi.org/10.1109/ispass.2016.7482092
Publikováno v:
ISPASS
While many techniques have been shown to be successful at reducing the amount of on-chip network traffic, no studies have shown how close a combined approach would come to eliminating all unnecessary data traffic, nor have any studies provided insigh
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::41b4ee9c2f5724e8ffc79c40f0364675
https://doi.org/10.1109/ispass.2015.7095798
https://doi.org/10.1109/ispass.2015.7095798
Publikováno v:
ASPLOS
Recent work has shown that disciplined shared-memory programming models that provide deterministic-by-default semantics can simplify both parallel software and hardware. Specifically, the DeNovo hardware system has shown that the software guarantees
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0dbbb9098b10cc5e42e0143f7b76a69d
https://doi.org/10.1145/2451116.2451119
https://doi.org/10.1145/2451116.2451119
Autor:
Nicholas P. Carter, Hyojin Sung, Robert Smolinski, Ching-Tsun Chou, Rakesh Komuravelli, Byn Choi, Nima Honarmand, Sarita V. Adve, Vikram Adve
Publikováno v:
PACT
For parallelism to become tractable for mass programmers, shared-memory languages and environments must evolve to enforce disciplined practices that ban "wild shared-memory behaviors;'' e.g., unstructured parallelism, arbitrary data races, and ubiqui
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::011abc705b4310b147127c623e441c4b
https://doi.org/10.1109/pact.2011.21
https://doi.org/10.1109/pact.2011.21
Autor:
Danny Dig, Mohsen Vakilian, Stephen T. Heumann, Robert L. Bocchino, Vikram Adve, Jeffrey Overbey, Hyojin Sung, Rakesh Komuravelli, Patrick Simmons, Sarita V. Adve
Publikováno v:
OOPSLA
Today's shared-memory parallel programming models are complex and error-prone.While many parallel programs are intended to be deterministic, unanticipated thread interleavings can lead to subtle bugs and nondeterministic semantics. In this paper, we
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9fd1c8f0b054487e070a3aca8f515657
https://doi.org/10.1145/1640089.1640097
https://doi.org/10.1145/1640089.1640097