Data Center Switch for Load Balanced Fat-Trees
- PDF / 1,709,119 Bytes
- 15 Pages / 595.276 x 790.866 pts Page_size
- 8 Downloads / 163 Views
Data Center Switch for Load Balanced Fat-Trees Wei-Chih Lai · Ching-Te Chiu
Received: 20 June 2012 / Revised: 3 October 2012 / Accepted: 15 October 2012 / Published online: 16 November 2012 © Springer Science+Business Media New York 2012
Abstract With the growing of cloud computing, the need of computing power no longer can be satisfied with a few powerful servers or small scale parallel computer systems. More and more servers are connected together as a data center network. Then, fault tolerance becomes an import issue when building a massive data center network. Currently, many researches focus on building fat-tree data center networks. In this paper, we propose a load balanced fat-tree architecture with uniform mapping connection patterns to provide higher fault tolerance capability for heavy traffic load networks. Two fault tolerated 4 × 4 banyan type switch designs are introduced to improve the fault tolerance capability of fat-tree networks. Finally, fault tolerance capability evaluations of link or switch faults in fattree network are given to support our idea, and a 4 × 4 banyan type switch IC is demonstrated as the commodity switch for building the fault tolerant fat-tree data center networks. The 4 × 4 banyan type switch IC is fabricated in 90 nm CMOS technology, and the maximum operation rate of the IC is 5.8 Gbps per channel or 23.2 Gbps total data rate with only 23 ps peak-to-peak jitter. Keywords Data center network · Scalability · Load balance · Fault tolerance · Fat-Tree · Load balanced Birkhoff–von Neumann switch
W.-C. Lai · C.-T. Chiu (B) Department of Computer Science, National Tsing Hua University, Hsinchu, Taiwan e-mail: [email protected]
1 Introduction Recently data center networks have become popular since cloud computing and data center service were announced. There are various cloud services and applications nowadays, such as Amazon Elastic Compute Cloud (Amazon cloud service), iGoogle (Google web 2.0 application), and many on-line anti-virus applications. With the growing of cloud computing, the need of computing power no longer can be satisfied with a few powerful servers or small scale parallel computer systems. One possible solution is to bring hundreds of thousands of servers together and connect them into a data center network. According to Cisco Data Center Infrastructure 2.5 Design Guide [1], the core-aggregation-access threetier tree architecture is adopted since tree topology is simple and easy to build. However, the bandwidth of interconnections near the core tier is huge so that the tree architecture needs numerous high-end switches and routers. The fat-tree architecture is recommended by Al-Fares et al. [2] to solve the interconnection capacity issue, and they also build a 27,648-node fat-tree network by using only commodity switches to support their idea. The fat-tree is a special type of tree that the number of interconnections increases from leaves to root switches for separating traffic into these interconnections. Building data center networks using fat-tree topology
Data Loading...
