Abstract
RPC is a communication technology which has been widely used in distributed systems. It has been employed as an essential component of distributed systems. However, the performance of traditional RPC technology will be seriously decreased in high-speed network based cluster system. The main reason is that the network of cluster has many obvious features such as high bandwidth, low latency, and high reliability, etc, which are different from normal network environment. Therefore, how to improve the performance of RPC technology in cluster system has caught increasing attention.
In this paper, we have carefully studied the traditional RPC technology, which suggest that the decreasing of cluster network performance is mainly caused by the serialization/deserialization process of RPC technology. Thus we proposed a dedicated serialization/deserialization scheme which can run on homogeneous cluster system. This scheme can well improve the performance of cluster network by reducing the number of date copy operations of RPC protocol. We have evaluated our improved scheme in our real-world cluster system. And our evaluation results show that our scheme can significantly promote performance of bandwidth by up to 43% in our cluster system when the size of transmitted data block is large.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Birrell, A.D., Nelson, B.J.: Implementing Remote Procedure Call. ACM Transactions on Computer Systems 2(1) (February 1984)
Buyya, R.: High performance cluster computing. Posts & Telecom Press (2002)
Chen, H., Shi, L., Sun, J.: VMRPC: A High Efficiency and Light Weight RPC System for Virtual Machines. In: 8th International Workshop on Quality of Service
David, D., Clark, V., Jacobson, J.: An Analysis of TCP processing overhead. IEEE Communications (June 1989)
Douglas, E., Comer, D.L.: Internetworking with TCP/IP. Client-Server Programming and Applications, vol. 3. Publishing House of Electronics Industry (2008)
Eisler, M.: XDR: External Data Representation Standard. RFC 4506 (May 2006)
http://docs.oracle.com/cd/E19683-01/816-1435/oncintro-3/index.html
Nagle, D., Serenyi, D., Matthews, A.: The Panasas ActiveScale Storage Cluster: Delivering Scalable High Bandwidth Storage. In: Proceedings of the 2004 ACM/IEEE Conference on Supercomputing (2004)
Braam, P.: The Lustre Storage Architecture (2004), http://www.lustre.org/docs/lustre.pdf
Richard Stevens, W.: UNIX Network Programming, 2nd edn. Interprocess Communications, vol. 2. Posts & Telecom Press (2009)
Sun Microsystems, Inc. LUSTRETM NETWORKING High-Performance Features and Flexible Support for a Wide Array of Networks, White Paper (November 2008)
Andrew, S., Van Steen, M.: Distributed systems: principles and paradigms. Tsinghua University Press (2008)
Thurlow, R.: RPC: Remote Procedure Call Protocol Specification Version 2, RFC5531 (May 2009)
Wan, Y., Feng, D., et al.: An In-depth Analysis of TCP and RDMA Performance on Modern Server Platform. In: The 7th IEEE International Conference on Networking, Architecture, and Storage
Welch, B., Unangst, M., et al.: Scalable Performance of the Panasas Parallel File System. In: 6th USENIX Conference on File and Storage Technologies (FAST 2008) (2008)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Wan, Y., Feng, D., Wang, F., Zhu, T. (2013). A Dedicated Serialization Scheme in Homogeneous Cluster RPC Communication. In: Park, J.J.(.H., Arabnia, H.R., Kim, C., Shi, W., Gil, JM. (eds) Grid and Pervasive Computing. GPC 2013. Lecture Notes in Computer Science, vol 7861. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-38027-3_43
Download citation
DOI: https://doi.org/10.1007/978-3-642-38027-3_43
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-38026-6
Online ISBN: 978-3-642-38027-3
eBook Packages: Computer ScienceComputer Science (R0)