By Topic

Bandwidth Management with a Reconfigurable Data Cache

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
Pradeep Nalabalapu ; Ambarella Corp., Sunnyvale, CA, USA ; Sass, R.

With ever larger FPGA devices, hardware engineers are increasingly relying on automated tools to generate complex designs. However, relatively little attention has focused on automatically generating components of the memory hierarchy. Conventional cache research (despite its extensive study) rarely offers designs that map well to FPGAs. Here we propose an approach that uses compiler technology to analyze an application's predominant array access patterns and then generates a data cache customized for the application. The generic Reconfigurable Data Cache component and the technique used to automatically configure it are described. To demonstrate the feasibility of the proposed approach, a prototype has been implemented. We use the convolution as a representative multimedia operation, and show the benefit of the Reconfigurable Data Cache. Even though the computational structure for convolution is easy to generate automatically (from high-level source code), the resulting design alone is memory-bound and not faster than a comparable microprocessor. However, with the addition of the customized Reconfigurable Data Cache, the resulting system runs 5× faster and outperforms the reference microprocessor.

Published in:

Parallel and Distributed Processing Symposium, 2005. Proceedings. 19th IEEE International

Date of Conference:

04-08 April 2005