Zobrazeno 1 - 10
of 15
pro vyhledávání: '"DONGUP KWON"'
Autor:
Suyeon Hur, Seongmin Na, Dongup Kwon, Joonsung Kim, Andrew Boutros, Eriko Nurvitadhi, Jangwoo Kim
Publikováno v:
ACM Transactions on Architecture and Code Optimization. 20:1-24
Deep neural networks (DNNs) have become key solutions in the natural language processing (NLP) domain. However, the existing accelerators customized for their narrow target models cannot support diverse NLP models. Therefore, naively running complex
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::cb549db15f166174430bdb3d775bfe17
https://doi.org/10.1145/3564606
https://doi.org/10.1145/3564606
SmartFVM: A Fast, Flexible, and Scalable Hardware-based Virtualization for Commodity Storage Devices
Publikováno v:
ACM Transactions on Storage. 18:1-27
A computational storage device incorporating a computation unit inside or near its storage unit is a highly promising technology to maximize a storage server’s performance. However, to apply such computational storage devices and take their full po
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::00ca89a60ea7bfaac201adf5d01ebe66
https://doi.org/10.1145/3511213
https://doi.org/10.1145/3511213
Publikováno v:
IEEE Symposium on Security and Privacy
Security bugs in CPUs have critical security impacts to all the computation related hardware and software components as it is the core of the computation. In spite of the fact that architecture and security communities have explored a vast number of
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::85faa668689db79290e608813d775050
https://doi.org/10.1109/sp40001.2021.00103
https://doi.org/10.1109/sp40001.2021.00103
Publikováno v:
DAC
The increasing size of recurrent neural networks (RNNs) makes it hard to meet the growing demand for real-time AI services. For low-latency RNN serving, FPGA-based accelerators can leverage specialized architectures with optimized dataflow. However,
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::2144985703ab89292df0df93cab2088a
https://doi.org/10.1109/dac18072.2020.9218528
https://doi.org/10.1109/dac18072.2020.9218528
Publikováno v:
ISCA
A cost-effective multi-tenant neural network execution is becoming one of the most important design goals for modern neural network accelerators. For example, as emerging AI services consist of many heterogeneous neural network executions, a cloud pr
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ec7c30b30ed391cb05f028520e58d94c
https://doi.org/10.1109/isca45697.2020.00081
https://doi.org/10.1109/isca45697.2020.00081
Publikováno v:
IEEE Computer Architecture Letters. 17:47-50
SSD arrays are becoming popular in modern storage servers as a primary storage, and they aim to reduce the high cost of the devices by performing inline deduplications. Unfortunately, existing software-based inline deduplications cannot achieve the d
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::54c071f676dd58591549cba2d1365505
https://doi.org/10.1109/lca.2017.2753258
https://doi.org/10.1109/lca.2017.2753258
Autor:
Dongup Kwon, Ali Jafari, Abirami Prabhakaran, Pranavi Appana, Prerna Budhkar, Mishali Naik, Andrew Boutros, Eriko Nurvitadhi, Karthik Gururaj, Sheffield David B
Publikováno v:
FPT
We present a CPU server with multiple FPGAs that is purely software-programmable by a unified framework to enable flexible implementation of modern real-life complex AI that scales to large model size (100M+ parameters), while delivering real-time in
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::62f01f35ebe460002865dcaac0013aaf
https://doi.org/10.1109/icfpt47387.2019.00054
https://doi.org/10.1109/icfpt47387.2019.00054
Autor:
Bogdan Pasca, Dongup Kwon, Sergey Gribok, Gregory K. Chen, Eriko Nurvitadhi, Jaewoong Sim, Knag Phil, Martin Langhammer, Ram Krishnamurthy, Phillip Tomson, Debbie Marr, Aravind Dasu, Sumbul Huseyin Ekin, Ali Jafari, Raghavan Kumar, Andrew Boutros
Publikováno v:
FCCM
Interactive intelligent services, such as smart web search, are important datacenter workloads. They rely on dataintensive deep learning (DL) algorithms with strict latency constraints and thus require balancing both data movement and compute capabil
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::8424fcfb0c47be38b3c9b06d624d3d87
https://doi.org/10.1109/fccm.2019.00035
https://doi.org/10.1109/fccm.2019.00035
Autor:
Eriko Nurvitadhi, Raghavan Kumar, Martin Langhammer, Ali Jafari, Gregory K. Chen, Jaewoong Sim, Phillip Tomson, Sergey Gribok, Debbie Marr, Ram Krishnamurthy, Aravind Dasu, Knag Phil, Andrew Boutros, Bogdan Pasca, Dongup Kwon, Sumbul Huseyin Ekin
Publikováno v:
FPGA
Interactive intelligent services (e.g., smart web search) are becoming essential datacenter workloads. They rely on data-intensive artificial intelligence (AI) algorithms that do not use batch computation due to their tight latency constraints. Since
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ec019e5424875aa387716e89b6152bec
https://doi.org/10.1145/3289602.3293943
https://doi.org/10.1145/3289602.3293943
Publikováno v:
HPCA
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::b85274a29f0f52a79658000c0de91a44
https://doi.org/10.1109/hpca.2019.00025
https://doi.org/10.1109/hpca.2019.00025