Presentation is loading. Please wait.

Presentation is loading. Please wait.

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.

Similar presentations


Presentation on theme: "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."— Presentation transcript:

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 Im Kookmin University

2 Introduction To 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

3 Introduction Three type memory in GPU 1.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.

4 Introduction A new way of forwarding table lookup mechanism,which exploits the constant memory so that we can speed up the table lookup procedure.

5 Fast Forwarding Table Lookup Maintain the whole forwarding table in the global memory of GPU. Create indices in the constant memory(64KB).

6 Fast Forwarding Table Lookup Currently 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.

7 Fast Forwarding Table Lookup

8

9

10

11

12 Search:13.1.9.11

13 Performance Evaluation

14


Download ppt "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."

Similar presentations


Ads by Google