Computer and storage technology is evolving rapidly. Today, cloud computing is enabling the consolidation of traditional IT functions with entirely new capabilities. For example, many large-scale data centers are now providing traditional IT services along with new data analytics services.

Hence, these large-scale data centers require highly efficient server and storage systems. Traditional CPU technology limits performance, as the use of frequency scaling as a way to increase performance has ended. The end of frequency scaling has caused a shift to multicore processing. However, multicore processing has diminishing returns in terms of increasing true application performance due to limits in I/O and memory bandwidth.

Intel® FPGAs can be used to accelerate the performance of large-scale data systems. Intel FPGAs enable higher speed data processing by providing customized high-bandwidth, low-latency connections to network and storage systems. In addition,Intel FPGAs provide compression, data filtering, and algorithmic acceleration.

With the Intel® FPGA SDK for OpenCL™, you can now rapidly develop acceleration solutions for computer and storage systems. The Intel FPGA SDK for OpenCL enables even software developers to easily design with FPGAs by allowing them to utilize a high-level programming language for developing acceleration functions.

OpenCL and the OpenCL logo are trademarks of Apple Inc. used by permission by Khronos.

Cloud Computing

As cloud computing demands expand annually, FPGAs bring to the table offload capabilities, at a hardware level, for significant improvements in overall system performance.

Overview

Search

Big Data

 

 

 

Machine Learning

Machine learning applications continue to develop and increase in scope every year as artificial intelligence penetrates the data center. In this rapidly expanding field, reprogrammable FPGAs allow for the continual implementation of the newest algorithms and neural network topologies. 

Machine Learning/CNN

 

 

Storage Applications

Today's data centers require highly efficient storage systems to enable peak performance. Intel FPGAs offer high-speed data transaction solutions and storage processing solutions designed to alleviate I/O bottlenecks and to improve storage network performance.

Overview                         Networked Storage & RoCE

NAND Flash                     NVMe

Offload Engines

Enterprise

Implementations with FPGA in Enterprise applications accelerate the workload on design solutions include Oil & Gas, Genomics, Video Transcoding and more.

Video Transcoding

Financial / HPC

In the financial sector, software-based algorithms and analytics can be offloaded onto FPGA hardware for reductions in latencies. In high-performance computing, parallel tasks that may bog down CPU performance can be offloaded to the FPGA, reducing CPU overhead and system latency.

Financial Acceleration

Financial Offload