Abstract
Hosts with several, possibly heterogeneous and/or multicore, processors provide new challenges and opportunities to accelerate applications with high communications bandwidth requirements. Many opportunities to scale these network applications with the increase in the link bandwidths are related to the exploitation of the available parallelism provided by the presence of several processing cores in the servers, not only for computing the workload of the user application but also for decreasing the overhead associated to the network interface and the system software.
This paper analyses some alternatives to distribute the network interface among the different cores available in the host. These alternatives include the use of heterogeneous multicore processors, such as network processors, and the exploitation of the affinity between the network interface and the location (proximity to the memories where the different data structures are stored) and characteristics of the processing architecture. The proposed distributed network interfaces provide improvements in throughput and latency that have been experimentally evaluated by using an intrusion prevention system, which constitutes a useful network function.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Catanzaro B et al. (2010) Ubiquitous parallel computing from Berkeley, Illinois, and Stanford. IEEE Micro 41–55 (March/April)
Kumar R et al (2005) Heterogeneous chip multiprocessors. IEEE Comput 38(11):32–38
Hill M, Marty M (2008) Amdahl’s law in the multicore era. IEEE Comput 41(6):33–38
Willmann P, Shafer J, Carr D, Menon A, Rixner S, Cox AL, Zwaenepoel W (2007) Concurrent direct network access for virtual machine monitors. In: Proc intl symposium on high-performance computer architecture, February, 2007
Rixner S (2008) Network virtualization: breaking the performance barrier. ACM Queue (January/February): 36–52
Balaji P, Feng W, Panda DK (2006) Bridging the Ethernet-Ethernot performance gap. IEEE Micro (May–June): 24–40
Shivam P, Chase, JS (2003) On the elusive benefits of protocol offload. In: SIGCOMM’03 workshop on network-I/O convergence: experience, lessons, implications (NICELI), August, 2003
Regnier G et al (2004) TCP onloading for data center servers. IEEE Comput 37(11):48–58
Wun B, Crowley P (2006) Network I/O acceleration in heterogeneous multicore processors. In: Proceedings of the 14th annual symposium on high performance interconnects (hot interconnects), August, 2006
Ortiz A, Ortega J, Díaz AF, Prieto A (2009) A new offloaded/onloaded network interface for high performance communication. In: 17th euromicro international conference on parallel, distributed and network-based processing. PDP 2009, February, 2008
Shalev L, Marhervaks V, Machulsky Z, Biran G, Satran J, Ben-Yehuda M, Shimony I (2006) Loosely coupled TCP acceleration architecture. In: Proceedings of the 14th IEEE symposium on high-performance interconnects (HOTI)
Nahum EM, Yates DJ, Kurose JF, Towsley D (1994) Performance issues in parallelized network protocols. In: Proc of the operating systems design and implementation, pp 125–137
Kim H, Pai VS, Rixner S (2003) Exploiting task-level concurrency in a programmable network interface. In: Proc of the ACM PPoPP’03
Shivam P, Wyckoff P, Panda D (2002) Can user-level protocols take advantage of multi-CPU NICs? In: Proc intl parallel and distributed processing symp (IPDPS’02), April, 2002, pp 64–69
Narayanaswamy G, Balaji P, Feng W (2007) An analysis of 10-gigabit Ethernet protocol staks in multicore environments. In: 15th IEEE symp on high-performance interconects (HOTI’07), pp 109–116
Foong A, Fung J, Newell D (2004) An in-depth analysis of the impact of processor affinity on network performance. In: Proceedings of the 12th IEEE international conference on networks
Salehi J, Kurose J, Towsley D (1996) The effectiveness of affinity-based scheduling in multiprocessor network protocol processing. IEEE/ACM Trans Netw 4(4):516–530
Jan H, Jin H-W (2009) MiAMI: multi-core aware processor affinity for TCP/IP over multiple network interfaces. In: 17th IEEE symposium on high performance interconnects. HOTI
de Bruijn W, Bos H (2008) Model-T: Rethinking the OS for terabits speeds. In: Workshop on high-speed networks (HSN2008), INFOCOM’2008
Mudigonda J, Vin HM, Yavatkar R (2005) Overcoming the memory wall in packet processing: hammers or ladders. In: Proc of the ACM ANCS’05.
Irqbalance daemon (2011) GNU General Public License (GPL) version 2. http://irqbalance.org/
Johnson EJ, Kunze AR (2003) IXP2400/2800 programming. The complete microengine coding guide. Intel Press, Santa Clara
Netronome: NFE-i800 Network Acceleration Card (2006) www.netronome.com/pages/acceleration-cards
Love R (2005) Linux kernel development, 2nd edn. Novell Press, Atlanta
Netfilter framework (2011) www.netfilter.org
Benvenuti C (2006) Understanding Linux network internals, 1st edn. O’Reilly Media, Sebastopol
Binkert N et al (2006) Integrated network interfaces for high-bandwidth TCP/IP. In: Proc of the ASPLOS conference, December, 2006
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Cascón, P., Ortiz, A., Ortega, J. et al. Accelerating network applications by distributed interfaces on heterogeneous multiprocessor architectures. J Supercomput 58, 302–313 (2011). https://doi.org/10.1007/s11227-011-0586-4
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-011-0586-4