Scalable and Cost-Effective Interconnection of Data-Center Servers Using Dual Server Ports

The goal of data-center networking is to interconnect a large number of server machines with low equipment cost while providing high network capacity and high bisection width. It is well understood that the current practice where servers are connected by a tree hierarchy of network switches cannot meet these requirements. In this paper, we explore a new server-interconnection structure. We observe that the commodity server machines used in today's data centers usually come with two built-in Ethernet ports, one for network connection and the other left for backup purposes. We believe that if both ports are actively used in network connections, we can build a scalable, cost-effective interconnection structure without either the expensive higher-level large switches or any additional hardware on servers. We design such a networking structure called FiConn. Although the server node degree is only 2 in this structure, we have proven that FiConn is highly scalable to encompass hundreds of thousands of servers with low diameter and high bisection width. We have developed a low-overhead traffic-aware routing mechanism to improve effective link utilization based on dynamic traffic state. We have also proposed how to incrementally deploy FiConn.

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

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

[3]  Sajal K. Das,et al.  Book Review: Introduction to Parallel Algorithms and Architectures : Arrays, Trees, Hypercubes by F. T. Leighton (Morgan Kauffman Pub, 1992) , 1992, SIGA.

[4]  W. Dally,et al.  Route packets, not wires: on-chip interconnection networks , 2001, Proceedings of the 38th Design Automation Conference (IEEE Cat. No.01CH37232).

[5]  Frank Thomson Leighton Introduction to parallel algorithms and architectures: arrays , 1992 .

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

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

[8]  Laxmi N. Bhuyan,et al.  A general class of processor interconnection strategies , 1982, ISCA '82.

[9]  GhemawatSanjay,et al.  The Google file system , 2003 .

[10]  Behrooz Parhami,et al.  Introduction to Parallel Processing: Algorithms and Architectures , 1999 .

[11]  Dmitri Loguinov,et al.  Graph-theoretic analysis of structured peer-to-peer systems: routing distances and fault resilience , 2003, IEEE/ACM Transactions on Networking.

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

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

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

[15]  Yuan Yu,et al.  Dryad: distributed data-parallel programs from sequential building blocks , 2007, EuroSys '07.

[16]  Sanjay Ghemawat,et al.  MapReduce: Simplified Data Processing on Large Clusters , 2004, OSDI.

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