1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K....
-
date post
20-Dec-2015 -
Category
Documents
-
view
213 -
download
0
Transcript of 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K....
![Page 1: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/1.jpg)
1
Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines
Author:Weirong Jiang Viktor K. Prasanna
Publisher:ACM 2008
Presenter:Po Ting Huang
Date:2010/3/03
![Page 2: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/2.jpg)
2
Outline
Introduction Architecture Front End Back End Memory Balancing Trie Partitioning
Subtrie-to-Pipeline Mapping Node-to-Stage Mapping
Performance
![Page 3: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/3.jpg)
3
Introduction
To balance the memory distribution across stages, several novel pipeline architectures have been proposed [2, 11, 6],but none of them can perfectly balanced memory distribution over stages
Ring→throughput degradation, packet blocking during a route update.
CAMP→delay variation, packet blocking during a route update. OLP→the first several stages in OLP may not be balanced The “memory wall” tends to impede the performance improvem
ent of a single pipeline architecture. Thus it becomes necessary to employ multiple pipelines paralle
l search architecture to speed IP lookup.
![Page 4: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/4.jpg)
4
The architecture consists of multiple bidirectional linear pipelines where each pipeline stores part of a routing table
Two mapping schemes are proposed to balance the memory distribution over different pipelines as well as across different stages in each pipeline.
![Page 5: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/5.jpg)
5
Architecture
![Page 6: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/6.jpg)
6
Front End
receives packets and dispatches the packets to different pipelines
Cache: most recently searched IP addresses and their next-hop information.
DIT: stores the relationship between the subtries and the pipeline entrances (destination index table)
Scheduler: determines which pipeline entrance the packet is routed to
![Page 7: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/7.jpg)
7
Back End
processes the packets and outputs the retrieved next-hop information.
multi-port queue:to tolerate the access conflict
output delay queue: When the packet exits the pipeline, it may be delayed to be output so that the intra-flow packet order is preserved
![Page 8: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/8.jpg)
8
Trie Partitioning
initial stride(I): The number of initial bits to be used A larger I:more small subtries, but result in prefix duplication Prefix duplication: results in memory inefficiency and may incre
ase the update cost
I=2
![Page 9: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/9.jpg)
9
We study the prefix length distribution based on four representative routing tables collected from [17]
Following sections, we pick I = 12 as default.
![Page 10: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/10.jpg)
10
Subtrie-to-Pipeline Mapping
Si denotes the set of subtries contained in the ith pipeline,i=1 to p
K is the number of subtries
Ti is the ith subtrie,i=1 to k
![Page 11: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/11.jpg)
11
Subtrie-to-Pipeline Mapping
![Page 12: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/12.jpg)
12
Node-to-Stage Mapping
Constraint: If node A is an ancestor of node B in a trie, then A must be mapped to a stage preceding the stage to which B is mapped.
The main ideas are to allow (1) two subtries to be mapped onto different directions (2) two trie nodes on the same trie level to be mapped onto different sta
ges.
![Page 13: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/13.jpg)
13
Node-to-Stage Mapping
several heuristics to select the subtries to be inverted
1.Largest leaf
2.Least height
3.Largest leaf per height
4.Least average depth per leaf
IFR denotes the inversion factor IFR=0→no subtrie is inverted
IFR close to the pipeline depth→all subtries are inverted
![Page 14: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/14.jpg)
14
Node-to-Stage Mapping
→
IFR=1
Use Least average depth per leaf heuristics
![Page 15: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/15.jpg)
15
↓
![Page 16: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/16.jpg)
16
The priority of a trie node is defined as forward subtrie →The priority of a trie node is definedas its height
reverse subtrie→its depth
The node whose priority is equal to the number of the remaining
stages is regarded as a critical node.
Node fields – Distance to child– Memory address of child
![Page 17: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/17.jpg)
17
Performance
I=12 P=4 H=25 IFR=4~8 Least average
depth per leaf
heuristics
![Page 18: 1 Multi-Terabit IP Lookup Using Parallel Bidirectional Pipelines Author: Weirong Jiang Viktor K. Prasanna Publisher: ACM 2008 Presenter: Po Ting Huang.](https://reader035.fdocuments.us/reader035/viewer/2022062714/56649d435503460f94a1e907/html5/thumbnails/18.jpg)
18
Performance
Memory: 1.8 MB– (13+5)*2^13*25*4=14.75Mb=1.8MB– 18 KB/stage
18.75 G packets / sec– 7.5 PPC*2.5GHz=18.75G packets/sec=6.0Tbps(packet size
=40bytes)
Power consumption: 0.2 W / IP lookup– 0.008*25=0.2W