Abstract
InfiniBand Architecture(IBA) adopts a mechanism of Virtual Lanes(VL) to support various QoS, improve link utilization and avoid deadlock. IBA packets are placed in a buffer corresponding to each VL and are scheduled by a VL arbiter. Since the number of the packets through each VL may be different, the throughput can be improved by allocating buffers according to the packet characteristics. In this paper, a new buffer allocation scheme using VL arbitration table is proposed and the superiority of the network performance is shown through the simulation results.
This work is partly supported by the IT System Research Center, Korea Electronics Technology Institute(KETI).
Corresponding author
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
Yuan-Hau Yeh, Chen-Yi Lee, “Cost-effective VLSI architectures and buffer size optimization for full-search block matching algorithms,” Very Large Scale Integration (VLSI) Systems, IEEE Transactions on, Vol. 7, pp. 345–358, Sep, 1999.
Tsang-Ling Sheu, Kuo-Fu Lee, “A multiple-class buffer allocation scheme for ATM networks with VBR traffic,” Information Networking, 1998. (ICOIN-12) Proceedings., Twelfth International Conference on, pp. 144–149, Jan, 1998.
InfiniBand Trade Association, http://www.infinibandta.org
InfiniBand Trade Association, “InfiniBand Architecture Release 1.0a,” Vol. 1, June 19, 2001.
Pfister, G..F., “Aspects of the InfiniBand Architecture,” Cluster Computing, 2001, Proceedings 2001, IEEE International Conference on, pp. 369–371, 2001.
Alfaro, F.J., Sanchez, J.L., Duato, J., “A strategy to manage time sensitive traffic in InfiniBand,” Parallel and Distributed Processing Symposium., Proceedings International, IPDPS 2002, Abstracts and CD-ROM, pp. 167–174, 2002.
Sancho, J.C., Flich, J., Robles, A., Lopez, P., and Duato, J., “Analyzing the influence of virtual lanes on the performance of InfiniBand networks,” Parallel and Distributed Processing Symposium., Proceedings International, IPDPS 2002, Abstracts and CD-ROM, pp. 166–175, 2002.
Alfaro, F.J., Sanchez, J.L., Orozco, L., and Duato, J., “Performance evaluation of VBR traffic in InfiniBand,” Electrical and Computer Engineering, 2002. IEEE CCECE 2002. Canadian Conference on, Vol. 3, pp. 1532–1537, 2002.
Joe Pelissier, “Providing Quality of Service over InfiniBand architecture fabrics,” Proce-edings of the 8th Symposium on Hot Interconnects, Aug, 2000.
Chunk Semeria, “Supporting Differentiated Service Classes: Queue Scheduling Disci-plines,” Juniper Networks white paper, Dec, 2001.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Yi, H., Park, S., Kim, M., Jeon, K. (2003). An Efficient Buffer Allocation Technique for Virtual Lanes in InfiniBand Networks. In: Chung, CW., Kim, CK., Kim, W., Ling, TW., Song, KH. (eds) Web and Communication Technologies and Internet-Related Social Issues — HSI 2003. HSI 2003. Lecture Notes in Computer Science, vol 2713. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45036-X_28
Download citation
DOI: https://doi.org/10.1007/3-540-45036-X_28
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-40456-9
Online ISBN: 978-3-540-45036-8
eBook Packages: Springer Book Archive