Intel® FPGA SDK for OpenCL™ Pro Edition: Best Practices Guide

ID 683521
Date 12/13/2021
Public

A newer version of this document is available. Customers should click here to go to the newest version.

Document Table of Contents

8.2.3. Optimizing for Two or More Banks of Global Memory

The Intel® FPGA SDK for OpenCL™ Offline Compiler automatically creates a global memory interconnect designed to deliver most of the available global memory bandwidth from the BSP to the kernel system.

The throughput can be saturated using read-only, write-only, or mixed read/write traffic. By default, traffic is interleaved across all available banks. If you find that the throughput is insufficient, Intel® recommends using the -no-interleaving option.