There is growing interest in network processor technologies capable of processing packets at line rates. In this paper, we present the design, implementation and evaluation of NetBind, a high performance, flexible and scalable binding tool for dynamically constructing data paths in network processor-based routers. The methodology that underpins NetBind balances the flexibility of network programmability against the need to process and forward packets at line speeds. Data paths constructed using NetBind seamlessly share the resources of the same network processor. We compare the performance of NetBind to the MicroACE system developed by Intel to support binding between software components running on Intel IXP1200 network processors. We evaluate these alternative approaches in terms of their binding overhead, and discuss how this can affect the forwarding performance of IPv4 data paths running on IXP1200 network processor-based routers. We show that NetBind provides better performance in comparison to MicroACE with smaller binding overhead. The NetBind source code described and evaluated in this paper is freely available on the Web (comet.columbia.edu/genesis/netbind) for experimentation. .

No abstract

No abstract

There is growing interest in network processor technologies capable of processing packets at line rates. Network processors are likely to be an integral part of next generation high-speed router and switch architectures, replacing the application-specific integrated circuits (ASICs) that are used in routers today. In this paper, we present the design, implementation and evaluation of NetBind, a high-performance, flexible and scalable binding tool for dynamically constructing data paths in network processor-based routers. The methodology that underpins NetBind balances the flexibility of network programmability against the need to process and forward packets at line speeds. To support the dynamic binding of components with the minimum addition of instructions in the critical path, NetBind modifies the machine language code of components at run time. To support fast data path composition, NetBind reduces the number of binding operations required for constructing data paths to a minimum set so that binding latencies are comparable with packet forwarding times. Data paths constructed using NetBind seamlessly share the resources of the same network processor. Resources are assigned during the binding process. We compare the performance of NetBind to the MicroACE system developed by Intel and show that NetBind provides better performance in comparison to MicroACE with smaller binding overhead. The NetBind source code described and evaluated in this paper is freely available on the Web (http://www.comet.columbia.edu/genesis/netbind) for experimentation.There has been a growing interest in network processor technologies [1-3] that can support softwarebased implementations of the critical path while processing packets at high speeds. Network processors employ multiple processing units to offer high packet-processing throughput. We believe that introducing programmability into network processor-based routers is an important area of research that has not been fully addressed as yet. The difficulty stems from the fact network processor-based routers need to forward minimum size packets at line rates and yet support modular and extensible data paths. Typically, the higher the line rate supported by a network processor-based router the smaller the set of instructions that can be executed in the critical path.Data path modularity and extensibility requires the dynamic binding between independently developed packet processing components. While code modularity and extensibility are supported by programming environments running in host processors (e.g. high-level programming language compilers and linkers), such capability cannot be easily offered in the network. Traditional techniques for realizing code binding (e.g. insertion of code stubs or indirection through function tables), introduce some overhead when applied to network processors in terms of additional instructions in the critical path. This overhead can be significant in some network processors. One solution to this problem is to optimize the code produced by a binding ...

No abstract