Home   >   CSC-OpenAccess Library   >    Manuscript Information
Full Text Available

(453.31KB)
This is an Open Access publication published under CSC-OpenAccess Policy.
Performance Review of Zero Copy Techniques
Jia Song, Jim Alves-Foss
Pages - 256 - 268     |    Revised - 15-07-2012     |    Published - 10-08-2012
Volume - 6   Issue - 4    |    Publication Date - August 2012  Table of Contents
MORE INFORMATION
KEYWORDS
Zero Copy, Network Security, Security/Perofrmance Tradeoff
ABSTRACT
E-government and corporate servers will require higher performance and security as usage increases. Zero copy refers to a collection of techniques which reduce the number of copies of blocks of data in order to make data transfer more efficient. By avoiding redundant data copies, the consumption of memory and CPU resources are reduced, thereby improving performance of the server. To eliminate costly data copies between user space and kernel space or between two buffers in the kernel space, various schemes are used, such as memory remapping, shared buffers, and hardware support. However, the advantages are sometimes overestimated and new security issues arise. This paper describes different approaches to implementing zero copy and evaluates these methods for their performance and security considerations, to help when evaluating these techniques for use in e-government applications
CITED BY (4)  
1 Morrell, C., Ransbottom, J. S., Marchany, R., & Tront, J. G. (2014, December). Scaling IPv6 address bindings in support of a moving target defense. In Internet Technology and Secured Transactions (ICITST), 2014 9th International Conference for (pp. 440-445). IEEE.
2 Victor, A. (2014). Splice Zerocopy System optimization using dual pipe schemes (Doctoral dissertation, TELKOM UNIVERSITY).
3 Wei, J., Zhihong, T., Chao, C., & Bei, G. (2014). Bottleneck analysis for data acquisition in high-speed network traffic monitoring. Communications, China, 11(1), 110-118.
4 Eisfeld, A. (2014). Entwurf und Analyse von Konzepten zur effizienten Datenübertragung von Grafikrendering-Befehlen auf eingebetteten Systemen.
1 Google Scholar
2 CiteSeerX
3 refSeek
4 Scribd
5 slideshare
6 PdfSR
1 E. Brose. “ZeroCopy: Techniques, Benefits and Pitfalls,” http://kbs.cs.tuberlin. de/teaching/ws2005/htos/papers/streams-zcpy.pdf, 2005, [last accessed May 30, 2012]
2 C. Huang, C. Chen, S. Yu, S. Hsu, and C. Lin. “Accelerate in-line packet processing using fast queue,” in Proc. 2010 IEEE Region 10 Conference, 2010, pp. 1048–1052.
3 D. Stancevic. “Zero Copy I: User-Mode Perspective,” Linux Journal, vol. 2003 no 105, Jan. 2003, pp. 3.
4 P. Halvorsen, E. Jorde, K. Skevik, V. Goebel, T. Plagemann, “Performance Tradeoffs for Static Allocation of Zero-Copy Buffers,” in Proc. Euromicro Conference, 2002, pp. 138-143.
5 M. Chiang and Y. Li, “LyraNET: A Zero-Copy TCP/IP Protocol Stack for Embedded Operating Systems,” in Proc. IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, Aug. 2005, pp. 123-128.
6 Linux Programmer's Manual, http://www.unix.com/man-page/Linux/2/sendfile/ Feb. 2010.
7 J.Corbet, A.Rubini, and G.Kroah-Hartman. “Memory Mapping and DMA” in Linux Device Drivers, third edition, O’REILLY, Feb, 2005, pp.450.
8 Linux Programmer’s Manual, http://www.unix.com/man-page/Linux/2/splice/ Sep. 2009.
9 K. Fall and J. Pasquale, “Exploiting In-Kernel Data Paths to Improve I/O Throughput and CPU Availability,” in Proc. Winter 1993 USENIX Conference, 1993, pp. 327-333.
10 H.K.J. Chu. “Zero-copy TCP in Solaris,” Proc. USENIX Annual Technical Conference, 1996, pp. 253–264.
11 S. Yamagiwa, K. Aoki, and K.Wada. “Active zero-copy: A performance study of nondeterministic messaging,” in Proc. International Symposium on Parallel and Distributed Computing, 2005, pp. 325–332.
Miss Jia Song
University of Idaho - United States of America
song3202@vandals.uidaho.edu
Dr. Jim Alves-Foss
University of Idaho - United States of America