Relating the bisection width of dual-port, server-centric datacenter networks and the solution of edge isoperimetric problems in graphs

Abstract Stellar datacenter networks are a recent generic construction designed to transform a base-graph into a dual-port, server-centric datacenter network. We prove that the S-bisection width of any stellar datacenter network can be obtained from the solution of isoperimetric problems on the base-graph, provided that the base-graph is regular. We extend previous research on the stellar datacenter networks GQ⁎, instantiated with generalized hypercubes, and show that with respect to S-bisection width, GQ⁎ performs well in comparison with the dual-port datacenter network FiConn. Our work develops a strong combinatorial link between graph bisection width and throughput metrics for stellar datacenter networks.

[1]  Iain A. Stewart,et al.  An Optimal Single-Path Routing Algorithm in the Datacenter Network DPillar , 2015, IEEE Transactions on Parallel and Distributed Systems.

[2]  Haitao Wu,et al.  Scalable and Cost-Effective Interconnection of Data-Center Servers Using Dual Server Ports , 2011, IEEE/ACM Transactions on Networking.

[3]  Iain A. Stewart,et al.  On the Mathematics of Data Centre Network Topologies , 2015, FCT.

[4]  I. Stoica,et al.  A Cost Comparison of Data Center Network Architectures , 2010 .

[5]  Iain A. Stewart,et al.  Improved Routing in the Data Centre Networks HCN and BCN , 2014, 2014 Second International Symposium on Computing and Networking.

[6]  William J. Dally,et al.  Principles and Practices of Interconnection Networks , 2004 .

[7]  Amin Vahdat,et al.  Helios: a hybrid electrical/optical switch architecture for modular data centers , 2010, SIGCOMM '10.

[8]  Yunhao Liu,et al.  Expandable and Cost-Effective Network Structures for Data Centers Using Dual-Port Servers , 2013, IEEE Transactions on Computers.

[9]  Ion Stoica,et al.  A cost comparison of datacenter network architectures , 2010, CoNEXT.

[10]  Torsten Hoefler,et al.  Slim Fly: A Cost Effective Low-Diameter Network Topology , 2014, SC14: International Conference for High Performance Computing, Networking, Storage and Analysis.

[11]  F. Leighton,et al.  Introduction to Parallel Algorithms and Architectures: Arrays, Trees, Hypercubes , 1991 .

[12]  Jung Ho Ahn,et al.  HyperX: topology, routing, and packaging of efficient large-scale networks , 2009, Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis.

[13]  John H. Lindsey,et al.  Assignment of Numbers to Vertices , 1964 .

[14]  Lei Shi,et al.  Dcell: a scalable and fault-tolerant network structure for data centers , 2008, SIGCOMM '08.

[15]  Jie Wu,et al.  On the design and analysis of Data Center Network architectures for interconnecting dual-port servers , 2014, IEEE INFOCOM 2014 - IEEE Conference on Computer Communications.

[16]  Frank Thomson Leighton,et al.  Graph bisection algorithms with good average case behavior , 1984, Comb..

[17]  William J. Dally,et al.  Technology-Driven, Highly-Scalable Dragonfly Topology , 2008, 2008 International Symposium on Computer Architecture.

[18]  David S. Johnson,et al.  Computers and Intractability: A Guide to the Theory of NP-Completeness , 1978 .

[19]  Koji Nakano.  Linear Layouts of Generalized Hypercubes , 1993, WG.

[20]  JOSEP DÍAZ,et al.  A survey of graph layout problems , 2002, CSUR.

[21]  L. H. Harper.  Optimal Assignments of Numbers to Vertices , 1964 .

[22]  Jie Wu,et al.  DCube: A family of network structures for containerized data centers using dual-port servers , 2014, Comput. Commun..

[23]  Reinhard Diestel,et al.  Graph Theory , 1997 .

[24]  Jason Lee,et al.  Generalized DCell Structure for Load-Balanced Data Center Networks , 2010, 2010 INFOCOM IEEE Conference on Computer Communications Workshops.

[25]  Haitao Wu,et al.  BCube: a high performance, server-centric network architecture for modular data centers , 2009, SIGCOMM '09.

[26]  Frank Thomson Leighton,et al.  A Framework for Solving VLSI Graph Layout Problems , 1983, J. Comput. Syst. Sci..

[27]  Dharma P. Agrawal,et al.  Generalized Hypercube and Hyperbus Structures for a Computer Network , 1984, IEEE Transactions on Computers.

[28]  Charles E. Leiserson,et al.  Fat-trees: Universal networks for hardware-efficient supercomputing , 1985, IEEE Transactions on Computers.

[29]  Ankit Singla,et al.  Jellyfish: Networking Data Centers Randomly , 2011, NSDI.

[30]  Antonio Fernández,et al.  Bisection (Band)Width of Product Networks with Application to Data Centers , 2014, IEEE Trans. Parallel Distributed Syst..

[31]  Amin Vahdat,et al.  A scalable, commodity data center network architecture , 2008, SIGCOMM '08.

[32]  Sudhakar Yalamanchili,et al.  Adaptive routing in generalized hypercube architectures , 1991, Proceedings of the Third IEEE Symposium on Parallel and Distributed Processing.

[33]  Antony Rowstron,et al.  Symbiotic routing in future data centers , 2010, SIGCOMM 2010.

[34]  Ratko V. Tomic.  Optimal networks from error correcting codes , 2013, Architectures for Networking and Communications Systems.

[35]  Lixin Gao,et al.  DPillar: Dual-port server interconnection network for large scale data centers , 2012, Comput. Networks.

[36]  Iain A. Stewart,et al.  The stellar transformation: From interconnection networks to datacenter networks , 2015, Comput. Networks.

[37]  S. Bezrukov.  Edge Isoperimetric Problems on Graphs , 2007 .

[38]  Himanshu Shah,et al.  FireFly , 2014, SIGCOMM.

[39]  Sangeetha Abdu Jyothi,et al.  Measuring and Understanding Throughput of Network Topologies , 2014, SC16: International Conference for High Performance Computing, Networking, Storage and Analysis.

[40]  C. Thomborson,et al.  A Complexity Theory for VLSI , 1980 .

[41]  Burkhard Monien,et al.  Approximation Algorithms for Multilevel Graph Partitioning , 2007, Handbook of Approximation Algorithms and Metaheuristics.

[42]  William J. Dally,et al.  Flattened butterfly: a cost-efficient topology for high-radix networks , 2007, ISCA '07.