Presentation is loading. Please wait.

Presentation is loading. Please wait.

CSE 8383 - Advanced Computer Architecture Week-11 April 1, 2004 engr.smu.edu/~rewini/8383.

Similar presentations


Presentation on theme: "CSE 8383 - Advanced Computer Architecture Week-11 April 1, 2004 engr.smu.edu/~rewini/8383."— Presentation transcript:

1 CSE 8383 - Advanced Computer Architecture Week-11 April 1, 2004 engr.smu.edu/~rewini/8383

2 Contents Message Passing (Distributed Memory) Systems Static Networks Dynamic Networks

3 MIMD Distributed Memory Systems Interconnection Networks MMMM PPPP

4 Distributed Memory Multiple address spaces Communication via send & receive Synchronization via message passing

5 Interconnection Network Several IN classification criteria Several IN classification criteria Mode of operation: synchronous vs. Asynchronous Control strategy: centralized vs. decentralized Switching techniques: circuit vs. packet switching Topology: static vs. dynamic

6 Interconnection Network Taxonomy Interconnection Network Static Dynamic Bus-basedSwitch-based 1-D2-DHC SingleMultiple SSMS Crossbar

7 Static Interconnection Networks Direct links, which are fixed once built Fixed Connections, unidirectional or bi- directional Fully Connected (Completely Connected Network (CCN)) Limited Connection Network (LCN).

8 Dynamic Interconnection Networks Communication patterns are based on program demands Connections are established on the fly during program execution Multistage Interconnection Network (MIN) and Crossbar

9 Static Network Topology Linear arrays Ring (Loop) networks Two-dimensional arrays Tree networks Cube network ….

10 Static Network Analysis Graph Representation Parameters Cost Degree Diameter Fault tolerance

11 Graph Review G = (V,E) -- V: nodes, E: edges Directed vs. Undirected Weighted Graphs Path, path length, shortest path Cycles, cyclic vs. acyclic Connectivity: connected, weakly connected, strongly connected, fully connected

12 Linear Array N nodes, N-1 edges Node Degree: Diameter: Cost: Fault Tolerance:

13 Ring N nodes, N edges Node Degree: Diameter: Cost: Fault Tolerance:

14 Chordal Ring N nodes, N edges Node Degree: Diameter: Cost: Fault Tolerance:

15 Barrwl Shifter Number of nodes N = 2 n Start with a ring Add extra edges from each node to those nodes having power of 2 distance i & j are connected if |j-I| = 2 r, r = 0, 1, 2, …, n-1

16 Barrel Shifter N = 16 Node Degree: Diameter: Cost: Fault Tolerance:

17 Tree and Star Node Degree: Diameter: Cost: Fault Tolerance:

18 Mesh and Torus Node Degree: Internal  4 Other  3, 2 Diameter: 2(n-1) N = n*n Node Degree: 4 Diameter: 2* floor(n/2)

19 Fully Connected 63 12 54 Node Degree: Diameter: Cost: Fault Tolerance:

20 Hypercubes N = 2 d d dimensions (d = log N) A cube with d dimensions is made out of 2 cubes of dimension d-1 Symmetric Degree, Diameter, Cost, Fault tolerance Node labeling – number of bits

21 Hypercubes d = 0d = 1d = 2d = 3 0 1 01 00 11 10 000 001 100110 111 011 101 010

22 Hypercubes 1110 1111 1010 1011 0110 0111 0010 0011 1101 1010 1000 1001 0100 0101 0010 0000 0001 S d = 4

23 Hypercube of dimension d N = 2 d d = log n Node degree = d Number of bits to label a node = d Diameter = d Number of edges = n*d/2 Hamming distance! Routing

24 Subcubes and Cube Fragmentation What is a subcube? Shared Environment Fragmentation Problem Is it Similar to something you know?

25 Cube Connected Cycles (CCC) k-cube  2 k nodes k-CCC from k-cube, replace each vertex of the k cube with a ring of k nodes K-CCC  k* 2 k nodes Degree, diameter  3, 2k Try it for 3-cube

26 K-ary n-Cube n = cube dimension K = # nodes along each dimension N = k n Wraparound Hupercube  binary n-cube Tours  k-ary 2-cube

27 Analysis and performance metrics static networks Performance characteristics of static networks NetworkDegree(d)Diameter(D)Cost(#link)SymmetryWorst delay Fully connected N-11N(N-1)/2Yes1 Linear Array2N-1 NoN Binary Tree3 2(  log 2 N  –1) N-1Nolog 2 N n-cubelog 2 N nN/2Yeslog 2 N 2D-Mesh42(n-1)2(N-n)No NN K-ary n-cube2n n  k/2  nNYesK x log 2 N

28 Dynamic Network Analysis Parameters: Cost: number of switches Delay: latency Blocking characteristics Fault tolerance

29 Switch Modules A x B switch module A inputs and B outputs In practice, A = B = power of 2 Each input is connected to one or more outputs (conflicts must be avoided) One-to-one (permutation) and one-to- many are allowed

30 Binary Switch 2x2 Switch Legitimate States = 4 Permutation Connections = 2


Download ppt "CSE 8383 - Advanced Computer Architecture Week-11 April 1, 2004 engr.smu.edu/~rewini/8383."

Similar presentations


Ads by Google