What is


?

A reliable, scalable, distributed acceleration system



Program against your FPGAs like it’s a single pool of accelerators

Coral is a framework that allows the distributed acceleration of large data sets across clusters of FPGA resources using simple programming models. It is designed to scale up from single devices to thousands of FPGAs, each offering local computation and storage. Rather than rely on hardware to deliver high-availability, the library itself is designed to detect and handle failures at the application layer, so delivering a highly-available service on top of a cluster of FPGAs, each of which may be prone to failures.




Features

Coral abstracts FPGA resources (device, memory), enabling fault-tolerant heterogeneous distributed systems to easily be built and run effectively.


Ease of Use

Write applications quickly in C++, Java, Scala and Python.
InAccel offers all the required high-level functions that make it easy to build and accelerate parallel apps. No need to modify your application to use an unfamiliar parallel programming language (like OpenCL). And you can use it interactively from the Scala and Python shells. It also allows seamless integration with Apache Spark.

Generality

Build your own repository of accelerators.
InAccel provides a stack of cores including Machine Learning techniques, Financial formulas, SQL functions and Cryptographic algorithms. You can combine all these libraries, along with your own ones, seamlessly in the same application.

Scalability

Industry proven to easily scale to unlimited FPGA resources.

Runs Anywhere

Runs on any FPGA platform (Xilinx, Intel), giving you the freedom to take full advantage of on-premises, or public Cloud (AWS, Alibaba, Nimbix, etc.) infrastructure.

Easy to Deploy

Launch a container with InAccel's Docker image or even deploy it as a daemonset on a Kubernetes cluster and enjoy acceleration services at the drop of a hat.





Download the FPGA manager brochure

slide


More Features

Coral provides an acceleration abstraction layer for applications, virtualizing, managing and monitoring the available FPGA resources.


Resource Management

Automatic resource configuration and task scheduling across entire FPGA clusters in private datacenters or public cloud environments.
Coral examines the state of the FPGAs and implements load-balancing policies across them, efficiently taking care of all the required device configurations and memory transfers.

Privacy / Isolation

Coral allows the secure sharing of the hardware resources among different users and multiple processes or threads.
First class isolation support for accelerator cores and FPGA memory.

Web UI

Built-in Web UI for viewing cluster state, navigating in memory objects and monitoring acceleration tasks.

Integrations
















Pricing




Community edition
Helping developers build, deploy, and run their accelerated applications.

FREE

APIs in Scala, Java, Python and C++
Up to 2 nodes / Unlimited FPGAs per node
Native CLI automation tool
Bitstream protection mechanism

Enterprise edition
Enabling teams to speed acceleration delivery with security, automation, advanced configuration and workflow.

300$ / Node / Month

All features from Community Edition plus:
Unlimited nodes
Web UI for monitoring your FPGAs' performance
Bitstream deployment pipelines
24x7 uptime support

Supported FPGA cards

Xilinx Alveo family

U50, U200, U250, U280

Intel PAC A10

D5005, N3000


Supported servers

Dell EMC

PowerEdge with FPGAs

HPE

ProLiant

IBM

Power Systems

EXXACT

BOSTON

Semptian

Microway

AWS

f1.2x, f1.4x, f1.16x

Alibaba

f1.2x, f1.4x, f1.14x

Alibaba

f3.2x, f3.4x, f3.8x, f3.16x

Huawei

FACS

Baidu

Nimbix

Alveo

Bittware

Terabox 1400B