An Evaluation Architecture for a Network Coprocessor
نویسندگان
چکیده
This paper outlines research currently being conducted by the Cal Poly Intelligent Network Interface Card (CiNIC) project to develop a network coprocessor. The purpose of this coprocessor is to free the host machine from its network processing duties as well as to allow for additional functionality such as hardware-based firewalling or quality of service (QoS) support. We provide an overview of the current CiNIC architecture as well as an introduction to and evaluation of the next generation CiNIC architecture. Our evaluation consisted of analyzing the performance and capabilities of an FPGA processor in order to determine whether it will meet our future development requirements. The FPGA’s performance was tested by timing the execution of the uClinux TCP/IP stack during send operations. The processor’s capabilities were tested by adding custom logic to the system and interfacing it with the uClinux operating system. We determined that both the performance and flexibility of the FPGA make it an ideal next generation CiNIC architecture.
منابع مشابه
An OpenSHMEM Implementation for the Adapteva Epiphany Coprocessor
This paper reports the implementation and performance evaluation of the OpenSHMEM 1.3 specification for the Adapteva Epiphany architecture within the Parallella single-board computer. The Epiphany architecture exhibits massive many-core scalability with a physically compact 2D array of RISC CPU cores and a fast network-on-chip (NoC). While fully capable of MPMD execution, the physical topology ...
متن کاملAn FPGA Based Scheduling Coprocessor for Dynamic Priority Scheduling in Hard Real-Time Systems
In this paper we present a scheduling coprocessor device for uniprocessor computer systems running a real-time operating system (RTOS). The coprocessor shortens the scheduling time of the operating system by performing dynamic priority computation for all tasks in parallel and making a task selection according to these priorities at a higher speed than a software solution would do. This paper s...
متن کاملOn Network CoProcessors for Scalable, Predictable Media Services
This paper presents the embedded realization and experimental evaluation of a media stream scheduler on Network Interface (NI) CoProcessor boards. When using media frames as scheduling units, the scheduler is able to operate in real-time on streams traversing the CoProcessor, resulting in its ability to stream video to remote clients at real-time rates. The contributions of this paper are its d...
متن کاملLocality-Conscious Load Balancing: Connectionist Architectural Support
Traditionally, in distributed memory architectures, locality maintenance and load balancing are seen as user level activities involving compiler and runtime system support in software. Such software solutions require an explicit phase of execution, requiring the application to suspend its activities. This paper presents the rst (to our knowledge) architecture-level scheme for extracting localit...
متن کاملA configurable Classification Engine for Polymorphous Chip Architecture
The ever-increasing demands for bandwidth requirement, faster IP forwarding, efficient and effective firewall and flexible differentiated services has resulted in the evolution of sophisticated Network Processor Units (NPUs). We describe a novel approach to implement a pipelined, configurable IPv6 and IPv4 coprocessor for a Network Processor Unit. The coprocessor is capable of providing Forward...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2002