Coral FPGA manager 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 multiple 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.
Scale your applications to multiple FPGAs on the same server or to multiple servers without changing your code.
InAccel dispatch the accelerated functions to the FPGA cluster automatically.
InAccel manager allows seamless sharing of the FPGA resources. from:
– Multiple threads
– Multiple processes
– Multiple applications
– Multiple users
It handles the resource management and the scheduling of the accelerated function to the FPGA cluster automatically.
InAccel offers all the required high-level functions (APIs) that make it easy to accelerate your applications by offloading the most computation intensive tasks to the FPGAs.
No need to modify your application or to use an unfamiliar parallel programming language (like OpenCL).
And you can use it interactively from the Scala and Python shells.
InAccel decouples the bitstreams from the Software developer.
SW developers do not need any more to work with bitstreams.
FPGA Resource Manager automatically fetch the right bitstream based on the requested function invoking and handles the programming and scheduling of the FPGAs.
InAccel FPGA orchestrator comes with a unique Monitor tool that allows to get valuable insight on the performance of the kernels in the FPGAs.
The monitor tool can be used to get valuable information for the execution of the accelerators in the FPGAs.
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.
Deploy your accelerators from Apache Spark, Jupyter notebooks or event Binderhub. Instant deployment with zero configuration.
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.
InAccel supports all the FPGA boards compatible with the OpenCL frameworks from Xilinx and Intel.