Abstract
This paper presents the implementation and performance of the HVIA-GE card, which is a hardware implementation of the Virtual Interface Architecture (VIA) based on Gigabit Ethernet. VIA is a user-level communication interface for high performance PC clustering. The HVIA-GE card is a 32-bit/33MHz PCI adapter containing an FPGA for the VIA Protocol Engine (VPE) and a Gigabit Ethernet chip set to construct a high performance physical network. HVIA-GE performs virtual-to-physical address translation, doorbell, and send/receive completion operations in hardware without kernel intervention. In particular, the Address Translation Table (ATT) is stored on the local memory of the HVIA-GE card, and the VPE efficiently controls the address translation process by directly accessing the ATT. As a result, the communication overhead during send/receive transactions is greatly reduced. Our experimental results show a minimum latency of 12.2 μs, and a maximum bandwidth of 96.3 MB/s. In terms of minimum latency, HVIA-GE performs 4.7 times and 9.7 times faster than M-VIA, a software implementation of VIA, and TCP/IP, respectively, over Gigabit Ethernet. In addition, the maximum bandwidth of HVIA-GE is 52% and 60% higher than M-VIA and TCP/IP, respectively.
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.
Similar content being viewed by others
References
Baker, M., Farrell, P.A., Ong, H., Scott, S.L.: VIA Communication Performance on a Gigabit Ethernet Cluster. In: Sakellariou, R., Keane, J.A., Gurd, J.R., Freeman, L. (eds.) Euro-Par 2001. LNCS, vol. 2150, p. 132. Springer, Heidelberg (2001)
Banikazemi, M., Liu, J., Kutlug, S., Ramakrishna, A., Sadayappan, P., Sah, H., Panda, D.K.: VIBe: A Micro-benchmark Suite for Evaluating Virtual Interface Architecture (VIA) Implementations. In: Int’l Parallel and Distributed Processing Symposium, IPDPS (2001)
Bozeman, P., Saphir, B.: A Modular High Performance Implementation of the Virtual Interface Architecture. In: Proc. of the 2nd Extreme Linux Workshop (1999)
Buonadonna, P., Geweke, A., Culler, D.E.: An Implementation and Analysis of the Virtual Interface Architecture. In: Proc. of the Supercomputing 1998 (1998)
Emulex Corporation: Hardware-based (ASIC) implementation of the Virtual Interface (VI) standard, http://www.emulex.com/products/legacy/index.html
ftp://ftp.compaq.com/pub/supportinformation/papers/tc000602wp.pdf
Ong, H., Farrell, P.A.: Performance Comparison of LAM/MPI, MPICH, and MVICH on a Linux Cluster connected by a Gigabit Ethernet Network. In: Proc. of 4th Annual Linux Showcase & Conference (2000)
Virtual Interface Architecture Specification, http://www.viarch.org/
Yoon, I.S., Chung, S.H., Lee, B., Kwon, H.C.: Implementation and Performance Evaluation of M-VIA on AceNIC Gigabit Ethernet Card. In: Kosch, H., Böszörményi, L., Hellwagner, H. (eds.) Euro-Par 2003. LNCS, vol. 2790, pp. 995–1000. Springer, Heidelberg (2003)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Park, S., Chung, SH., Yoon, IS., Jung, IH., Lee, S.M., Lee, B. (2004). HVIA-GE: A Hardware Implementation of Virtual Interface Architecture Based on Gigabit Ethernet. In: Aykanat, C., Dayar, T., Körpeoğlu, İ. (eds) Computer and Information Sciences - ISCIS 2004. ISCIS 2004. Lecture Notes in Computer Science, vol 3280. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30182-0_21
Download citation
DOI: https://doi.org/10.1007/978-3-540-30182-0_21
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-23526-2
Online ISBN: 978-3-540-30182-0
eBook Packages: Springer Book Archive