Abstract
With the explosive growth of Internet, the low efficiency of BGP has seriously influenced its usability. In this work, a TLP (thread level parallelism) approach with local speculation is proposed to improve the BGP performance. The thread partition is locally performed on each separated sub-module at route processing, and the speculation strategy is implemented to guarantee the memory consistency and sequential commit. Experiments on Intel Quad-core server show that this approach reaches an average speedup of 1.46 under single peer, multi-peers and route flapping. It is also shown that the packet throughput can be improved greatly under multiple sessions by taking advantage of TLP.
Similar content being viewed by others
References
Rekhter Y. A Border Gateway Protocol 4 (BGP-4), RFC 4271, 2006
Labovitz C, Malan G R, Jahanian F. Internet routing instability. IEEE/ACM Trans Net, 1998, 6(5): 515–558
Griffin T, Shepherd F B, Wilfong G. The stable paths problem and interdomain routing. IEEE/ACM Trans Net, 2002, 10(1): 232–234
Varadhan K, Govindan R, Estrin D. Persistent route oscillations in inter-domain routing. Com Net, 2000, 32(1): 1–16
Geoff Huston. The BGP Report for 2005, ISP Column, 2006, available at http://ispcolumn.isoc.org/2006-06/bgpupds.html
Agarwal S, Chuah C N, Bhattacharyya S, et al. Impact of bgp dynamics on router cpu utilization. In: PANM 2004, LNCS 3015. Berlin: Springer-Verlag, 2004. 278–288
Bjorkman M, Gunningberg P. Performance Modeling of Multiprocessor Implementations of Protocols. IEEE/ACM Trans Net, 1998, 6(3): 262–273
Maruyama M, Takahashi N. CORErouter-I: An experimental parallel IP router using a cluster of workstations. IEICE Trans Com, 1997, E80-B(10): 1407–1414
Xiao X P, Ni L M. Parallel routing table computation for scalable IP routers. In: PCCAA 1998, LNCS 1362. Berlin: Springer-Verlag, 1998. 145–158
Zhang X Z, Zhu P D, Lu X C. Fully-distributed and highly-parallelized implementation model of bgp4 based on clustered routers. In: ICN 2005, LNCS 3421. Berlin: Springer-Verlag, 2005. 433–441
Wu K, Wu J P, Xu K. A tree-based distributed model for BGP route processing. In: HPCC 2006, LNCS 4208. Berlin: Springer-Verlag, 2006. 119–128
Vtune TM: Intel Vtune Performance Analyzer for Linux. Version 9.0, Santa Clara(CA): Intel Corporation, 2007
Yang H H, Wong D F. Efficient network flow based min-cut balanced partitioning. IEEE Trans CAD, 1996, 15(12):1533–1540
Sarkar V, Hennessy J. Partitioning parallel programs for macro-dataflow. In: LISP and Functional Programming, Cambridge, Massachusetts, 1986. 202–211
Kwok Y K, Ahmad I. Static scheduling algorithms for allocating directed task graphs to multiprocessors. ACM Com Surg, 1999, 31(4): 406–471
Krishnan V, Torrellas J. Hardware and software support for speculative execution of sequential binaries on a chip-multiprocessor. In: ICS 1998, New York, NY, 1998. 85–92
Olukotun K, Hammond L, Willey M. Improving the performance of speculatively parallel applications on the Hydra CMP. In: ICS 1999, New York, NY, 1999. 21–30
Hou E S H, Ansari N, Ren H. A genetic algorithm for multiprocessor scheduling. IEEE Trans Par Dis Sys, 1994, 5(2):113–120
Kwok Y K, Ahmad I. Dynamic critical-path scheduling: An effective technique for allocating task graphs onto multiprocessors. IEEE Trans Par Dis Sys, 1996, 7(5): 506–521
Zhang W Z, Fang B X, Hu M Z, et al. Multisite co-allocation scheduling algorithms for parallel jobs in computing grid environments. Sci Chin Ser F-Inf Sci, 2006, 49(6): 906–926
Quagga: routing software suite. Version 0.99.9, Boston(MA): Free Software Foundation Inc., 2007
Author information
Authors and Affiliations
Corresponding author
Additional information
Supported by the National Basic Research Program of China (973 Program) (Grant No. 2003CB314802)
Rights and permissions
About this article
Cite this article
Gao, L., Gong, Z., Liu, Y. et al. A TLP approach for BGP based on local speculation. Sci. China Ser. F-Inf. Sci. 51, 1772–1784 (2008). https://doi.org/10.1007/s11432-008-0145-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11432-008-0145-7