Open Access. Powered by Scholars. Published by Universities.®

Engineering Commons

Open Access. Powered by Scholars. Published by Universities.®

Articles 1 - 2 of 2

Full-Text Articles in Engineering

An Open Source, Line Rate Datagram Protocol Facilitating Message Resiliency Over An Imperfect Channel, Christina Marie Smith Dec 2013

An Open Source, Line Rate Datagram Protocol Facilitating Message Resiliency Over An Imperfect Channel, Christina Marie Smith

Graduate Theses and Dissertations

Remote Direct Memory Access (RDMA) is the transfer of data into buffers between two compute nodes that does not require the involvement of a CPU or Operating System (OS). The idea is borrowed from Direct Memory Access (DMA) which allows memory within a compute node to be transferred without transiting through the CPU. RDMA is termed a zero-copy protocol as it eliminates the need to copy data between buffers within the protocol stack. Because of this and other features, RDMA promotes reliable, high throughput and low latency transfer for packet-switched networking. While the benefits of RMDA are well known and …


Memory Architecture Template For Fast Block Matching Algorithms On Field Programmable Gate Arrays, Shant Chandrakar Dec 2009

Memory Architecture Template For Fast Block Matching Algorithms On Field Programmable Gate Arrays, Shant Chandrakar

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Fast Block Matching (FBM) algorithms for video compression are well suited for acceleration using parallel data-path architectures on Field Programmable Gate Arrays (FPGAs). However, designing an efficient on-chip memory subsystem to provide the required throughput to this parallel data-path architecture is a complex problem. This thesis presents a memory architecture template that can be parameterized for a given FBM algorithm, number of parallel Processing Elements (PEs), and block size. The template can be parameterized with well known exploration techniques to design efficient on-chip memory subsystems. The memory subsystems are derived for two existing FBM algorithms and are implemented on a …