On the Effectiveness of Flow Aggregation in Improving Instruction Reuse in Network Processing Applications

Surendra, G.; Banerjee, S.; Nandy, S. K.
December 2003
International Journal of Parallel Programming;Dec2003, Vol. 31 Issue 6, p469
Academic Journal
Instruction Reuse is a microarchitectural technique that exploits dynamic instruction repetition to remove redundant computations at run-time. In this paper we examine instruction reuse of integer ALU and load instructions in network processing applications and attempt to answer the following questions: (1) How much of instruction repetition can be reused in packet processing applications?, (2) Can the temporal locality of network traffic be exploited to reduce interference in the Reuse Buffer and improve reuse? and (3) What is the effect of reuse on microarchitectural features such as resource contention and memory accesses? We use an execution driven simulation methodology to evaluate instruction reuse and find that for the benchmarks considered, 1 to 50% of the dynamic instructions are reused yielding performance improvement between 1 and 20%. To further improve reuse, a flow aggregation scheme as well as an architecture for exploiting the same is proposed. This scheme is mostly applicable to header processing applications and exploits temporal locality in packet data to uncover higher reuse. As a side effect, instruction reuse reduces memory traffic and improves performance.


Related Articles

  • Performance prediction and procurement in practice: assessing the suitability of commodity cluster components for wavefront codes. Hammond, S.D.; Mudalige, G.R.; Smith, J.A.; Davis, J.A.; Mills, A.B.; Jarvis, S.A.; Holt, J.; Miller, I.; Herdman, J.A.; Vadgama, A. // IET Software;Dec2009, Vol. 3 Issue 6, p509 

    The cost of state-of-the-art supercomputing resources makes each individual purchase a length and expensive process. Often each candidate architecture will need to be benchmarked using a variety of tools to assess likely performance. However, benchmarking alone only provides a limited insight...

  • Modelling user preferences and configuring services in B3G devices. Konstantinos Demestichas; Artemis Koutsorodi; Evgenia Adamopoulou; Michael Theologou // Wireless Networks (10220038);Oct2008, Vol. 14 Issue 5, p699 

    Abstract  This paper discusses a management architecture for devices operating in heterogeneous environments, that enables access network selection through terminal-controlled, preference-based mechanisms. In this domain two problems are identified, mathematically formulated and solved:...

  • Design and Development of HLA-based Operation-Maintenance Integration Training Simulation System. Xuanhuai Yang; Yuping Huo; Qing Tian; Xiue Zhang; Feng Gao // Applied Mechanics & Materials;2014, Issue 667-679, p1022 

    To meet the new training requirement caused by the changing of grid management mode, a training simulation system that fits to the new mode is proposed in this paper. Based on the HLA platform, an integrated simulation framework which could coordinate different simulation modules is designed....

  • An online pruning strategy for supervised ARTMAP-based neural networks. Shing Chiang Tan; Rao, M. V. C.; Chee Peng Lim // Neural Computing & Applications;2009, Vol. 18 Issue 4, p387 

    Identifying an appropriate architecture of an artificial neural network (ANN) for a given task is important because learning and generalisation of an ANN is affected by its structure. In this paper, an online pruning strategy is proposed to participate in the learning process of two constructive...

  • Measurement of the latency parameters of the Multi-BSP model: a multicore benchmarking approach. Savadi, Abdorreza; Deldari, Hossein // Journal of Supercomputing;Feb2014, Vol. 67 Issue 2, p565 

    Computer benchmarking is a common method for measuring the parameters of a computational model. It helps to measure the parameters of any computer. With the emergence of multicore computers, the evaluation of computers was brought under consideration. Since these types of computers can be viewed...

  • A SURVEY OF MODELS FOR COMPUTER NETWORKS MANAGEMENT. Kontoudis, Dimitris; Fouliras, Panayotis // International Journal of Computer Networks & Communications;May2014, Vol. 6 Issue 3, p157 

    The virtualization concept along with its underlying technologies has been warmly adopted in many fields of computer science. In this direction, network virtualization research has presented considerable results. In a parallel development, the convergence of two distinct worlds, communications...

  • Build it yourself. MacVittie, Don // Network World;10/25/2010, Vol. 27 Issue 20, p24 

    The article discusses how to optimize application software. It focuses on ensuring the secure, reliable and fast delivery of applications. It notes that optimization of applications addresses cost, flexibility, security and control issues. It examines the primary drivers to adopt application...

  • Beyond the box. Haigh, Ron // Network World;10/25/2010, Vol. 27 Issue 20, p24 

    The article offers information on managed application acceleration service. It discusses the disadvantages of technology including expensive cost, inflexible and complex. It notes that application acceleration technology needs to be fine-tuned with holistic view of the enterprise information...

  • Emerging Cloud Computing Paradigm. Zamani, Abu Sarwar; Akhtar, Mobin; Ahmad, Sultan // International Journal of Computer Science Issues (IJCSI);Jul2011, Vol. 8 Issue 4, p304 

    "Cloud" computing - a relatively recent term, defines the paths ahead in computer science world. Being built on decades of research it utilizes all recent achievements in virtualization, distributed computing, utility computing, and networking. It implies a service oriented architecture through...


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics