Advanced RDMA-Based Admission Control for Modern Data-Centers

Current data-centers employ admission control mechanism to maintain low response time and high throughput under overloaded scenarios. Existing mechanisms use internal (on the overloaded server) or external (on the front- end proxies) approaches. External admission control is preferred since it can be performed transparently without any modifications to the overloaded servers and global decisions can be made based on the load information of all the back-end servers. However, this external approach is bound to use TCP/IP communication protocol to get the load information from the back-end servers and rely on coarse-grained load monitoring due to the overheads associated with fine-grained monitoring. In this paper, we provide a fine-grained external admission control mechanism by leveraging the one-sided RDMA feature of modern interconnects and consequently provide response time guarantees and overload control in the multi-tiered data-center environment. Our design is implemented over InfiniBand-based clusters working in conjunction with Apache based servers. Experimental evaluations with single file, world cup and zipf traces show that our admission control can improve the response time by up to 28%, 17% and 23%, respectively, as compared to performing TCP/IP-based admission control. The improvement become 51%, 36% and 42%, respectively, when comparing to the base performance without any admission control. Further, our evaluations also show that RDMA-based admission control can provide better QoS guarantees than the TCP/IP-based admission control and no admission control approaches.

[1]  David E. Culler,et al.  SEDA: an architecture for well-conditioned, scalable internet services , 2001, SOSP.

[2]  Ludmila Cherkasova,et al.  Predictive admission control strategy for overloaded commercial Web server , 2000, Proceedings 8th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems (Cat. No.PR00728).

[3]  Erich M. Nahum,et al.  A method for transparent admission control and request scheduling in e-commerce web sites , 2004, WWW '04.

[4]  Per Anders Akersten,et al.  Availability and Maintainability , 2008 .

[5]  Nina Bhatti,et al.  Web server support for tiered services , 1999, IEEE Netw..

[6]  Amin Vahdat,et al.  Managing energy and server resources in hosting centers , 2001, SOSP.

[7]  William LeFebvre,et al.  CNN.com: Facing a World Crisis , 2001, LiSA.

[8]  George Kingsley Zipf,et al.  Human behavior and the principle of least effort , 1949 .

[9]  David E. Culler,et al.  USENIX Association Proceedings of USITS ’ 03 : 4 th USENIX Symposium on Internet Technologies and Systems , 2003 .

[10]  Hyun-Wook Jin,et al.  Exploiting RDMA operations for Providing Efficient Fine-Grained Resource Monitoring in Cluster-based Servers , 2006, 2006 IEEE International Conference on Cluster Computing.

[11]  Tao Yang,et al.  Selective early request termination for busy internet services , 2006, WWW '06.

[12]  Douglas M. Freimuth,et al.  Kernel Mechanisms for Service Differentiation in Overloaded Web Servers , 2001, USENIX Annual Technical Conference, General Track.

[13]  Michael Dahlin,et al.  Engineering web cache consistency , 2002, TOIT.