fast forwarding table lookup exploiting gpu memory architecture date:100/12/14 publisher:ieee ictc...
Post on 19-Dec-2015
212 views
TRANSCRIPT
![Page 1: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/1.jpg)
Fast Forwarding Table Lookup Exploiting GPU Memory Architecture
Date:100/12/14
Publisher:IEEE ICTC 2010
Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin Im
Kookmin University
![Page 2: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/2.jpg)
IntroductionTo speed up the software router processing
time,Packet Shader employers GPU to exploit the extensive parallelism that GPU can provide.
GPU contains a large number of small ALUs so that we can run the same number of threads as the number of ALUs
![Page 3: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/3.jpg)
IntroductionThree type memory in GPU1.Global Memory:highest bandwidth,large
size,but when two threads access the same location,the memory conflict should be resolved.
2.Constant Memory and Texture Memory do not show this problem and their access time is shorter,but size is too small to maintain forwarding table.
![Page 4: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/4.jpg)
IntroductionA new way of forwarding table lookup
mechanism,which exploits the constant memory so that we can speed up the table lookup procedure.
![Page 5: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/5.jpg)
Fast Forwarding Table LookupMaintain the whole forwarding table in the
global memory of GPU.Create indices in the constant memory(64KB).
![Page 6: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/6.jpg)
Fast Forwarding Table LookupCurrently the number of active BGP entries is
about 340,000.The number of prefixes with length of 16 or less
is 4.18%,24 is about 52%.Thus we expand all the prefixes of length less than 24 into multiple /24 prefixes.
The number of prefixes is still less than one million-20 bits.
![Page 7: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/7.jpg)
Fast Forwarding Table Lookup
![Page 8: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/8.jpg)
Fast Forwarding Table Lookup
![Page 9: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/9.jpg)
Fast Forwarding Table Lookup
![Page 10: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/10.jpg)
Fast Forwarding Table Lookup
![Page 11: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/11.jpg)
Fast Forwarding Table Lookup
![Page 12: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/12.jpg)
Fast Forwarding Table Lookup
Search:13.1.9.11
![Page 13: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/13.jpg)
Performance Evaluation
![Page 14: Fast Forwarding Table Lookup Exploiting GPU Memory Architecture Date:100/12/14 Publisher:IEEE ICTC 2010 Author:Youngjun Lee,Minseon Jeong,Sanghwan Lee,Eun-Jin](https://reader030.vdocument.in/reader030/viewer/2022032800/56649d2c5503460f94a02446/html5/thumbnails/14.jpg)
Performance Evaluation