Throughput-centric routing algorithm design

The increasing application space of interconnection networks now encompasses several applications, such as packet routing and I/O interconnect, where the throughput of a routing algorithm, not just its locality, becomes an important performance metric. We show that the problem of designing oblivious routing algorithms that have high worst-case or average-case throughput can be cast as a linear program. Globally optimal solutions to these optimization problems can be efficiently found, yielding provably good oblivious routing algorithms. Applying these techniques to k-ary 2-cube (tori) networks shows that previous routing algorithms sacrifice too much locality to achieve optimal worst-case throughput. This motivates the development of two new algorithms, IVAL and 2TURN, which improve locality to within 0.3% of optimal for an 8-ary 2-cube. Both algorithms have simple, deadlock-free implementations. Expanding the analysis of tori to average-case throughput reveals that there is a weak tradeoff between average-case and worst-case throughput. Specifically, both the IVAL and 2TURN algorithms developed for the worst-case also have good average-case throughput.

[1]  Bruce M. Maggs,et al.  Fast algorithms for bit-serial routing on a hypercube , 1990, SPAA '90.

[2]  Theodore R. Bashkow,et al.  A large scale, homogeneous, fully distributed parallel machine, I , 1977, ISCA '77.

[3]  Harald Räcke,et al.  Minimizing Congestion in General Networks , 2002, FOCS.

[4]  Allan Borodin,et al.  Adversarial queueing theory , 1996, STOC '96.

[5]  S. Lennart Johnsson,et al.  ROMM routing on mesh and torus networks , 1995, SPAA '95.

[6]  Allan Borodin,et al.  Routing, merging and sorting on parallel models of computation , 1982, STOC '82.

[7]  Dimitri P. Bertsekas,et al.  Nonlinear Programming , 1997 .

[8]  Yossi Azar,et al.  Throughput-competitive on-line routing , 1993, Proceedings of 1993 IEEE 34th Annual Foundations of Computer Science.

[9]  Ravindra K. Ahuja,et al.  Network Flows: Theory, Algorithms, and Applications , 1993 .

[10]  Baruch Awerbuch,et al.  Universal-stability results and performance bounds for greedy contention-resolution protocols , 2001, JACM.

[11]  Matthias Beck,et al.  The Ehrhart Polynomial of the Birkhoff Polytope , 2003, Discret. Comput. Geom..

[12]  William J. Dally,et al.  Worst-case Traffic for Oblivious Routing Functions , 2002, IEEE Computer Architecture Letters.

[13]  Christos Kaklamanis,et al.  Tight bounds for oblivious routing in the hypercube , 1990, SPAA '90.

[14]  Richard D. McBride,et al.  Progress Made in Solving the Multicommodity Flow Problem , 1998, SIAM J. Optim..

[15]  Dimitri P. Bertsekas,et al.  Data Networks , 1986 .

[16]  Francisco de Asis Ros Perán.  Routing to minimize the maximum congestion in a communication network , 1979 .

[17]  William J. Dally,et al.  GOAL: a load-balanced adaptive routing algorithm for torus networks , 2003, ISCA '03.

[18]  Allan Borodin,et al.  Adversarial queuing theory , 2001, JACM.

[19]  Lisa Fleischer,et al.  Approximating Fractional Multicommodity Flow Independent of the Number of Commodities , 2000, SIAM J. Discret. Math..

[20]  Ashish Goel,et al.  Source routing and scheduling in packet networks , 2005, JACM.

[21]  Farhad Shahrokhi,et al.  The maximum concurrent flow problem , 1990, JACM.

[22]  William J. Dally,et al.  Deadlock-Free Message Routing in Multiprocessor Interconnection Networks , 1987, IEEE Transactions on Computers.

[23]  Jochen Könemann,et al.  Faster and simpler algorithms for multicommodity flow and other fractional packing problems , 1998, Proceedings 39th Annual Symposium on Foundations of Computer Science (Cat. No.98CB36280).

[24]  Leslie G. Valiant,et al.  Universal schemes for parallel communication , 1981, STOC '81.

[25]  Luigi Fratta,et al.  The flow deviation method: An approach to store-and-forward communication network design , 1973, Networks.

[26]  David K. Smith.  Network Flows: Theory, Algorithms, and Applications , 1994 .

[27]  William J. Dally,et al.  Locality-preserving randomized oblivious routing on torus networks , 2002, SPAA '02.

[28]  William J. Dally,et al.  A delay model and speculative architecture for pipelined routers , 2001, Proceedings HPCA Seventh International Symposium on High-Performance Computer Architecture.