Achieving Real Data Throughput for an FPGA Co-Processor on Commodity Server Platforms
نویسندگان
چکیده
Significant investigation has taken place in the area of exploiting reconfigurable logic as a coprocessor in embedded and high-performance computer systems. This paper examines the problem of delivering high throughput data to the co-processor. Data is retrieved from magnetic storage and delivered to the coprocessor through datapaths available in inexpensive, commodity server platforms. Endto-end data throughput of 667 MB/s is achieved.
منابع مشابه
Achieving a Billion Requests Per Second Throughput on a Single Key-Value Store Server Platform via Full Stack Architecting
Distributed in-memory key-value stores (KVSs), such as memcached, have become a critical data serving layer in modern Internet-oriented datacenter infrastructure. Their performance and efficiency directly affect the QoS of web services and the efficiency of datacenters. Traditionally, these systems have had significant overheads from inefficient network processing, OS kernel involvement, and co...
متن کاملServerSwitch: A Programmable and High Performance Platform for Data Center Networks
As one of the fundamental infrastructures for cloud computing, data center networks (DCN) have recently been studied extensively. We currently use pure software-based systems, FPGA based platforms, e.g., NetFPGA, or OpenFlow switches, to implement and evaluate various DCN designs including topology design, control plane and routing, and congestion control. However, software-based approaches suf...
متن کاملMidas: An FPGA-based Architecture Simulator for Multiprocessors
We present Midas, an economical FPGA-based architecture simulator that allows rapid early design-space exploration of manycore systems. Midas models target-system timing and functionality separately, and it employs hostmultithreading for an efficient FPGA implementation. It is a high-throughput, cycle-accurate full-system simulator, capable of booting real operating systems. The Midas prototype...
متن کاملArchitecture Exploration for Tree-based Option Pricing Models MEng Final Year Project Report
This project explores the application of reconfigurable hardware and GPUs to the acceleration of financial computation using tree-based pricing models. Two parallel pipelined architectures have been developed for option valuation using binomial trees and trinomial trees, with support for concurrent evaluation of independent options to achieve high pricing throughput. Two highly optimised GPU im...
متن کاملDesign and Implementation of Digital Demodulator for Frequency Modulated CW Radar (RESEARCH NOTE)
Radar Signal Processing has been an interesting area of research for realization of programmable digital signal processor using VLSI design techniques. Digital Signal Processing (DSP) algorithms have been an integral design methodology for implementation of high speed application specific real-time systems especially for high resolution radar. CORDIC algorithm, in recent times, is turned out to...
متن کامل