ON MODELLING AND ANALYSIS OF RECEIVE LIVELOCK AND CPU UTILIZATION IN HIGH-SPEED NETWORKS

K. Salah and K. El-Badawi

References

  1. [1] K. Ramakrishnan, Performance consideration in designing net-work interfaces, IEEE Journal on Selected Areas in Commu-nications, 11(2), 1993, 203–219. doi:10.1109/49.215016
  2. [2] J. Mogul & K. Ramakrishnan, Eliminating receive livelock inan interrupt-driven kernel, ACM Trans. Computer Systems,15 (3), 1997, 217–252. doi:10.1145/263326.263335
  3. [3] R. Morris, E. Kohler, J. Jannotti, & M. Kaashoek, Clickmodular router, ACM Trans. Computer Systems, 8(3), 2000,263–297.
  4. [4] I. Kim, J. Moon, & H.Y. Yeom, Timer-based interrupt miti-gation for high performance packet processing, Proc. 5th Int.Conf. on High-Performance Computing in the Asia-PacificRegion, Gold Coast, Australia, 2001.168
  5. [5] A. Indiresan, A. Mehra, & K.G. Shin, Receive livelock elimi-nation via intelligent interface backoff, TCL Technical Report(Michigan, USA: University of Michigan, 1998).
  6. [6] P. Druschel, Operating system support for high-speed commu-nication, Communications of the ACM, 39(9), 1996, 41–51. doi:10.1145/234215.234470
  7. [7] P. Druschel & G. Banga, Lazy receive processing (LRP):A network subsystem architecture for server systems, Proc.2nd USENIX Symposium on Operating Systems Design andImplementation, 1996, 261–276. doi:10.1145/238721.238786
  8. [8] P. Shivan, P. Wyckoff, & D. Panda, EMP: Zero-copy OS-bypass NIC-driven gigabit ethernet message passing, Proc.Supercomputing Conf. 2001, Denver, Colorado, USA, 2001.
  9. [9] C. Dovrolis, B. Thayer, & P. Ramanathan, HIP: Hybridinterrupt-polling for the network interface, ACM OperatingSystems Reviews, 35(4), 2001, 50–60. doi:10.1145/506084.506089
  10. [10] Alteon WebSystems Inc., Singapore, Jumbo frames, available athttp://www.psc.edu/∼mathis/MTU/AlteonExtendedFrames_W0601.pdf
  11. [11] A. Gallatin, J. Chase, & K. Yocum, Trapeze/IP: TCP/IP atnear-gigabit speeds, Annual USENIX Technical Conf., Mon-terey, Canada, 1999, 109–119.
  12. [12] C. Traw & J. Smith, Hardware/software organization of a highperformance ATM host interface, IEEE Journal on SelectedAreas in Communications, 11(2), 1993, 240–253. doi:10.1109/49.215019
  13. [13] C. Traw & J. Smith, Giving applications access to Gb/snetworking, IEEE Network, 7(4), 1993, 44–52. doi:10.1109/65.224055
  14. [14] K. Salah & K. El-Badawi, Performance evaluation of interrupt-driven kernels in gigabit networks, Proc. IEEE Globecom 2003,San Francisco, 2003, 3953–3957.
  15. [15] W. Leland, M. Taqqu, W. Willinger, & D. Wilson, On theself-similar nature of ethernet traffic, IEEE/ACM Trans. Net-working, 2(1), 1994, 1–15. doi:10.1109/90.282603
  16. [16] J. Brustoloni & P. Steenkiste, Effects of buffering semantics onI/O performance, Proc. 2nd USENIX Symp. Operating SystemsDesign and Implementation, Seattle, WA, 1996, 277–291. doi:10.1145/238721.238787
  17. [17] Z. Ditta, G. Parulkar, & J. Cox, The APIC approach to highperformance network interface design: Protected DMA andother techniques, Proc. IEEE INFOCOM 1997, Kobe, Japan,1997, 179–187.
  18. [18] H. Keng & J. Chu, Zero-copy TCP in Solaris, Proc. of theUSENIX 1996 Annual Technical Conf., San Diego, California,1996, 253–264.
  19. [19] P. Steenkiste, A systematic approach to host interface designfor high-speed networks, IEEE Computer Magazine, 24(3),1994, 44–52.
  20. [20] 3Com Corporation, Gigabit Server Network Interface Cards7100xx Family, available at http://pdf.superwarehouse.com/specs/DS3COMBC109285.pdf
  21. [21] A. Law & W. Kelton, Simulation Modelling and Analysis(Singapore: McGraw-Hill, 2nd Edition, 1991).

Important Links:

Go Back