Abstract
Network processors can be categorized into two types, the coprocessors-centric model in which the data-plane is handled by coprocessors, and the core-centric model in which the core processes most of the data-plane packets yet offloading some tasks to coprocessors. While the former has been properly explored over various applications, researches regarding the latter remain limited. Based on the previous experience of prototyping the virtual private network (VPN) over the IXP425 network processor, this work aims to derive design implications for the core-centric model performing computational intensive applications. From system and IC vendors' perspectives, the continuous-time Markov chain and Petri net simulations are adopted to explore this architecture. Analytical results prove to be quite inline with those of the simulation and implementation. With subsequent investigation we find that appropriate process run lengths can improve the effective core utilization by 2.26 times, and by offloading the throughput boosts 7.5 times. The results also suggest single process programming since context switch overhead impacts considerably on the performance.
- Braun, T., Günter, M., Kasumi, M., and Khalil, I. 1999. Virtual private network architecture. Tech. rep. California Technical Institute. IAM-99-001.Google Scholar
- Clark, C., Lee, W., Schimmel, D., Contis, C. Koné, M., and Thomas, A. 2004. A hardware platform for network intrusion detection and prevention. In Proceedings of the 3rd Workshop on Network Processors and Applications (NP3), ACM, New York, NY.Google Scholar
- Comer, D. and Martynov, M. 2006. Building experimental virtual routers with network processors. In Proceedings of the 2nd International Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities (TRIDENTCOM). IEEE, Los Alamitos, CA.Google Scholar
- Crowley, P. and Baer, J.-L. 2002. A modeling framework for network processor systems. In Proceedings of the Network Processor Workshop in Conjunction with the 8th International Symposium on High Performance Computer Architecture (NP1). ACM, New York, NY.Google Scholar
- Davis, J.D., Fu, C., and Laudon, J. 2005. The RASE (rapid, accurate simulation environment) for chip multiprocessors. In Proceedings of the Workshop on Design, Architecture and Simulation of Chip Multiprocessors (dasCMP'05). ACM, New York, NY. Google Scholar
Digital Library
- Intel. IXP425 Network Processor. http://www.intel.com/design/network/products/npfamily/ixp425.htm.Google Scholar
- Intel. IXP2400 Network Processor. http://www.intel.com/design/network/products/npfamily/ixp2400.htm.Google Scholar
- Lekkas, P. C. 2003. Network Processors: Architectures, Protocols and Platforms (Telecom Engineering). McGraw-Hill Professional, New York, NY. Google Scholar
Digital Library
- Lin, Y.-D., Lin, Y.-N., Yang, S.-C., and Lin, Y.-S. 2003. DiffServ Edge Routers over Network Processors: Implementation and Evaluation. IEEE Network 17, 4, 28--34. Google Scholar
Digital Library
- Lin, Y.-N., Lin, C.-H., Lin, Y.-D., and Lai, Y.-C. 2005. VPN gateways over network processors: implementation and evaluation. In Proceedings of the 11th IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS'05). IEEE, Los Alamitos, CA. Google Scholar
Digital Library
- Lin, Y.-N., Chang, Y.-C., Lin, Y.-D., and Lai, Y.-C. Lai. 2007. Resource allocation in network processors for memory access intensive applications. J. Syst. Softw. 80, 7. Google Scholar
Digital Library
- Lu, J. and Wang, J. 2006. Analytical performance analysis of network-processor-based application designs. In Proceedings of the 15th International Conference on Computer Communications and Networks (IC3N'06). IEEE, Los Alamitos, CA, 33--39.Google Scholar
- Microsoft TechNet. http://www.microsoft.com/technet/prodtechnol/windows2000serv/reskit/core/fned_ana_trrf.mspx?mfr=true.Google Scholar
- Murata, T. 1989. Petri nets: Properties, analysis and applications. Proc. IEEE 77, 4.Google Scholar
Cross Ref
- Nussbaum, D., Fedorova, A., and Small, C. 2004. An overview of the Sam CMT simulator Kit. Tech. rep. Sun Microsystems. Google Scholar
Digital Library
- Ratzer, A. V., et al. 2003. CPN tools for editing, simulating, and analysing coloured Petri nets. In Proceedings of the International Conference on Applications and Theory of Petri Nets. Springer, Berlin, Germany. Google Scholar
Digital Library
- Smith, J. M. S. 1997. Application-Specific Integrated Circuits. Addison-Wesley, Upper Saddle River, NJ.Google Scholar
- Tan, Z., Lin, C., Yin, H., and Li, B. 2004. Optimization and benchmark of cryptographic algorithms on network processors. IEEE Micro 24, 5, 55--69. Google Scholar
Digital Library
- Wolf, T. and Franklin, M. K. 2006. Performance models for network processor design. IEEE Trans. Parall. Distrib. Syst. 17, 6, 548--561. Google Scholar
Digital Library
- Zuberek, W. M., Govindarajan, R., and Suciu, F. 1998. Timed colored Petri net models of distributed memory multithreaded multiprocessors. In Proceedings of Workshop on Practical Use of Colored Petri Nets and Design/CPN. ACM, New York, NY.Google Scholar
Index Terms
Modeling and analysis of core-centric network processors
Recommendations
Modeling and analysis of core-centric network processors
Network processors can be categorized into two types, the coprocessors-centric model in which data-plane is handled by coprocessors, and the core-centric model in which the core processes most of the data-plane packets yet offloading some tasks to ...
Enabling SIMT Execution Model on Homogeneous Multi-Core System
Single-instruction multiple-thread (SIMT) machine emerges as a primary computing device in high-perfor-mance computing, since the SIMT execution paradigm can exploit data-level parallelism effectively. This article explores the SIMT execution potential ...
Towards effective network algorithms on multi-core network processors
ANCS '08: Proceedings of the 4th ACM/IEEE Symposium on Architectures for Networking and Communications SystemsTo build high-performance network devices with holistic security protection, a large number of algorithms have been proposed. However, multi-core implementation of the existing algorithms suffers from three limitations: performance instability, data-...






Comments