Download presentation

1
**EE384y: Packet Switch Architectures**

Matchings, implementation and heuristics Nick McKeown Professor of Electrical Engineering and Computer Science, Stanford University Spring 2004

2
**Outline Finding a maximum match. What algorithms are used in practice?**

Maximum network flow problems Definitions and example Augmenting paths Maximum size/weight matchings as examples of maximum network flows Maximum size matching Complexity of maximum size matchings and maximum weight matchings What algorithms are used in practice? Maximal Matches Wavefront Arbiter (WFA) Parallel Iterative Matching (PIM) iSLIP Spring 2004

3
**Network Flows a c Source s Sink t b d**

10 1 Source s Sink t b d Let G = [V,E] be a directed graph with capacity cap(v,w) on edge [v,w]. A flow is an (integer) function, f, that is chosen for each edge so that We wish to maximize the flow allocation. Spring 2004

4
**A maximum network flow example By inspection**

10 1 Source s Sink t b d Step 1: Source s Sink t a c b d 10, 10 10 1 Flow is of size 10 Spring 2004

5
**A maximum network flow example**

Step 2: a c 10, 10 Source s Sink t 10, 10 1 10, 10 1 10, 1 b d 10, 1 1, 1 Source s Sink t a c b d 10, 10 10, 2 10, 9 1,1 1, 1 Maximum flow: Flow is of size 10+2 = 12 Not obvious Flow is of size 10+1 = 11 Spring 2004

6
**Ford-Fulkerson method of augmenting paths**

Set f(v,w) = -f(w,v) on all edges. Define a Residual Graph, R, in which res(v,w) = cap(v,w) – f(v,w) Find paths from s to t for which there is positive residue. Increase the flow along the paths to augment them by the minimum residue along the path. Keep augmenting paths until there are no more to augment. Spring 2004

7
**Example of Residual Graph**

c 10, 10 10, 10 1 s 10, 10 t 1 10 10 b d 1 Flow is of size 10 Residual Graph, R res(v,w) = cap(v,w) – f(v,w) a c 10 10 10 1 s t 1 10 10 b d 1 Augmenting path Spring 2004

8
**Example of Residual Graph**

Step 2: a c 10, 10 s t 10, 10 1 10, 10 1 10, 1 b d 10, 1 1, 1 Flow is of size 10+1 = 11 Residual Graph a c 10 s t 10 10 1 1 1 1 b d 1 9 9 Spring 2004

9
**Example of Residual Graph**

Step 3: a c 10, 9 s t 10, 10 1, 1 10, 10 1, 1 10, 2 b d 10, 2 1, 1 Flow is of size 10+2 = 12 1 Residual Graph a c 9 s t 10 10 1 1 2 2 b d 1 8 8 Spring 2004

10
**Complexity of network flow problems**

In general, it is possible to find a solution by considering at most |V|.|E| paths, by picking shortest augmenting path first. There are many variations, such as picking most augmenting path first. Spring 2004

11
**Outline Finding a maximum match. What algorithms are used in practice?**

Maximum network flow problems Definitions and example Augmenting paths Maximum size/weight matchings as examples of maximum network flows Maximum size matching Complexity of maximum size matchings and maximum weight matchings What algorithms are used in practice? Maximal Matches Wavefront Arbiter (WFA) Parallel Iterative Matching (PIM) iSLIP Spring 2004

12
**Finding a maximum size match**

B C D E F 1 2 3 4 5 6 How do we find the maximum size (weight) match? Spring 2004

13
**Network flows and bipartite matching**

1 B 2 Sink t Source s C 3 D 4 E 5 F 6 Finding a maximum size bipartite matching is equivalent to solving a network flow problem with capacities and flows of size 1. Spring 2004

14
**Network flows and bipartite matching Ford-Fulkerson method**

Residual Graph for first three paths: A 1 B 2 t s C 3 D 4 E 5 F 6 Spring 2004

15
**Network flows and bipartite matching**

Residual Graph for next two paths: A 1 B 2 t s C 3 D 4 E 5 F 6 Spring 2004

16
**Network flows and bipartite matching**

Residual Graph for augmenting path: A 1 B 2 t s C 3 D 4 E 5 F 6 Spring 2004

17
**Network flows and bipartite matching**

Residual Graph for last augmenting path: A 1 B 2 t s C 3 D 4 E 5 F 6 Note that the path augments the match: no input and output is removed from the match during the augmenting step. Spring 2004

18
**Network flows and bipartite matching**

Maximum flow graph: A 1 B 2 t s C 3 D 4 E 5 F 6 Spring 2004

19
**Network flows and bipartite matching**

Maximum Size Matching: A 1 B 2 C 3 D 4 E 5 F 6 Spring 2004

20
**Complexity of Maximum Matchings**

Maximum Size Matchings: Algorithm by Dinic O(N5/2) Maximum Weight Matchings Algorithm by Kuhn O(N3) In general: Hard to implement in hardware Slooooow. Spring 2004

21
**Outline Finding a maximum match. What algorithms are used in practice?**

Maximum network flow problems Definitions and example Augmenting paths Maximum size/weight matchings as examples of maximum network flows Maximum size matching Complexity of maximum size matchings and maximum weight matchings What algorithms are used in practice? Maximal Matches Wavefront Arbiter (WFA) Parallel Iterative Matching (PIM) iSLIP Spring 2004

22
Maximal Matching A maximal matching is one in which each edge is added one at a time, and is not later removed from the matching. i.e. no augmenting paths allowed (they remove edges added earlier). No input and output are left unnecessarily idle. Spring 2004

23
**Example of Maximal Size Matching**

B C D E F 1 2 3 4 5 6 A 1 B C D E F 2 3 4 5 6 A 1 B C D E F 2 3 4 5 6 Maximal Size Matching Maximum Size Matching Spring 2004

24
Maximal Matchings In general, maximal matching is simpler to implement, and has a faster running time. A maximal size matching is at least half the size of a maximum size matching. A maximal weight matching is defined in the obvious way. A maximal weight matching is at least half the weight of a maximum weight matching. Spring 2004

25
**Outline Finding a maximum match. What algorithms are used in practice?**

Maximum network flow problems Definitions and example Augmenting paths Maximum size/weight matchings as examples of maximum network flows Maximum size matching Complexity of maximum size matchings and maximum weight matchings What algorithms are used in practice? Maximal Matches Wavefront Arbiter (WFA) Parallel Iterative Matching (PIM) iSLIP Spring 2004

26
**Wave Front Arbiter (Tamir)**

Requests Match 1 1 1 1 2 2 2 2 3 3 3 3 4 4 4 4 Spring 2004

27
Wave Front Arbiter Requests Match Spring 2004

28
**Wave Front Arbiter Implementation**

Simple combinational logic blocks 1,1 1,2 1,3 1,4 2,1 2,2 2,3 2,4 3,1 3,2 3,3 3,4 4,1 4,2 4,3 4,4 Spring 2004

29
**Wave Front Arbiter Wrapped WFA (WWFA)**

N steps instead of 2N-1 Requests Match Spring 2004

30
**Wavefront Arbiters Properties**

Feed-forward (i.e. non-iterative) design lends itself to pipelining. Always finds maximal match. Usually requires mechanism to prevent Q11 from getting preferential service. In principle, can be distributed over multiple chips. Spring 2004

31
**Outline Finding a maximum match. What algorithms are used in practice?**

Maximum network flow problems Definitions and example Augmenting paths Maximum size/weight matchings as examples of maximum network flows Maximum size matching Complexity of maximum size matchings and maximum weight matchings What algorithms are used in practice? Maximal Matches Wavefront Arbiter (WFA) Parallel Iterative Matching (PIM) iSLIP Spring 2004

32
**Parallel Iterative Matching**

1 2 3 4 #1 #2 Iteration: uar selection uar selection 1 2 3 4 1 2 3 4 f2: Grant 1 2 3 4 f3: Accept/Match f1: Requests 1 2 3 4 1 2 3 4 Spring 2004

33
PIM Properties Guaranteed to find a maximal match in at most N iterations. In each phase, each input and output arbiter can make decisions independently. In general, will converge to a maximal match in < N iterations. How many iterations should we run? Spring 2004

34
**Parallel Iterative Matching Convergence Time**

Number of iterations to converge: Spring 2004

35
**Parallel Iterative Matching**

Spring 2004

36
**Parallel Iterative Matching**

PIM with a single iteration Spring 2004

37
**Parallel Iterative Matching**

PIM with 4 iterations Spring 2004

38
**Outline Finding a maximum match. What algorithms are used in practice?**

Maximum network flow problems Definitions and example Augmenting paths Maximum size/weight matchings as examples of maximum network flows Maximum size matching Complexity of maximum size matchings and maximum weight matchings What algorithms are used in practice? Maximal Matches Wavefront Arbiter (WFA) Parallel Iterative Matching (PIM) iSLIP Spring 2004

39
iSLIP 1 2 3 4 F2: Grant 1 2 3 4 1 2 3 4 F3: Accept/Match 1 2 3 4 #1 #2 F1: Requests 1 2 3 4 1 2 3 4 Spring 2004

40
iSLIP Operation Grant phase: Each output selects the requesting input at the pointer, or the next input in round-robin order. It only updates its pointer if the grant is accepted. Accept phase: Each input selects the granting output at the pointer, or the next output in round-robin order. Consequence: Under high load, grant pointers tend to move to unique values. Spring 2004

41
**iSLIP Properties Random under low load TDM under high load**

Lowest priority to MRU 1 iteration: fair to outputs Converges in at most N iterations. (On average, simulations suggest < log2N) Implementation: N priority encoders 100% throughput for uniform i.i.d. traffic. But…some pathological patterns can lead to low throughput. Spring 2004

42
iSLIP Spring 2004

43
iSLIP Spring 2004

44
**iSLIP Implementation Programmable Priority Encoder State Decision**

1 1 log2N Decision Grant Accept 2 2 N Grant Accept log2N N N N Grant Accept log2N Spring 2004

45
Maximal Matches Maximal matching algorithms are widely used in industry (PIM, iSLIP, WFA and others). PIM and iSLIP are rarely run to completion (i.e. they are sub-maximal). We will see shortly that a maximal match with a speedup of 2 is stable for non-uniform traffic. Spring 2004

Similar presentations

OK

Copyright © 2011, Elsevier Inc. All rights reserved. Chapter 6 Author: Julia Richards and R. Scott Hawley.

Copyright © 2011, Elsevier Inc. All rights reserved. Chapter 6 Author: Julia Richards and R. Scott Hawley.

© 2018 SlidePlayer.com Inc.

All rights reserved.

To make this website work, we log user data and share it with processors. To use this website, you must agree to our Privacy Policy, including cookie policy.

Ads by Google