Blog

Scaling Performance of DNN Inference in FPGA clusters using InAccel orchestrator

How Xilinx used InAccel Coral for the evaluation of Multi-FPGA deployment for DNN. The MLPerf evaluation on multi-FPGA accelerators has been done using the InAccel Coral orchestrator and resource manager.

GPUs vs FPGAs: Which one is better in DL and Data Centers applications

However there are several cases in the domain of deep learning that GPUs are considered more powerful than FPGAs. Then, why AMD decided to acquire Xilinx for $35 billion instead of further advancing its own GPUs?

Simplify AI Deployment at scale on FPGA clusters

InAccel provides a unique FPGA cluster manager that allows easy of deployment and scaling at production level for AI and ML applications. The FPGA manager can be used both for inference and training of ML models.

FPGA-accelerated ML on MindsDB's Lightwood

InAccel provides the Logistic Regression accelerator which can accelerate the training time of a machine learning model and it is available for Intel's Arria10 FPGAs, Amazon EC2 F1 instances and Xilinx's Alveo U200, U250 and U280 FPGA platforms. So we developed a LogisticRegression class that can be selected as the desired mixer to create the Lightwood Predictor.

What is missing for the success of the AI HW chips

The latest version of the Coral resource manager allows the software community to instantiate and utilize a cluster of AI hardware accelerators with the same easy as invoking typical software functions. InAccel’s Coral resource manager allows multiple applications to share and utilize a cluster of accelerators in the same node (server) without worrying about the scheduling, load balancing and the resource management of each accelerator.

Inaccel announces record-breaking speed on facial detection test using an FPGA cluster

InAccel, a world-pioneer in the domain of FPGA-based accelerators, today released the results of a test in which a cluster of eight FPGAs in a single server were used to achieve 1266 fps in a facial detection speed test.

Accelerated Face detection on a cluster of 8 Alveo FPGA cards

InAccel has released today an integrated framework that allows to utilize the power of an FPGA cluster for face detection. Specifically, InAccel has presented a demo in which a cluster of 8 FPGAs are used to provide up to 1700 fps (supporting up to 56 cameras with 30 fps in a single server).

Accelerated MLPerf inference on FPGA clusters: InAccel unique solution

Scaling out Inspur’s TF2 Deep Learning Inference engine

In this case we show how InAccel orchestrator can be integrated with TF2 to allow multi-tenant scalable deployment of TF2 to a cluster of Intel-based FPGAs. We show how InAccel’s orchestrator allows easy deployment, scaling, resource management, and task scheduling for FPGAs making it easier than ever, the deployment and the utilization of FPGA for Deep Learning Inference.

Hybrid, Heterogeneous auto-scalable FPGA deployment

Enterprise customers now have the option to adopt a hybrid deployment for their FPGA-accelerated applications. That means, that companies can purchase a small number of FPGAs deployed on-prem and if these FPGAs are overloaded, they can automatically scale it using the cloud resources. InAccel, using the unique platform-agnostic FPGA orchestration allows auto-scaling hybrid FPGA deployment in the most efficient way.

Accelerated Deep Learning Inference using FPGAs from your browser

Any user can now enjoy the speedup of the FPGA accelerators from their browser. In the DL example we show how users can enjoy much faster ResNet50 inference from the same Keras python notebook with zero code changes.

Deploying FPGAs over OpenShift for accelerated Machine Learning

InAccel, a pioneer on the domain of application accelerations, has released today the easiest way to utilize FPGA-based accelerators over OpenShift. InAccel has developed a unique FPGA orchestrator that abstracts away the FPGA resources making much easier the deployment, scaling and automated resource management of the accelerators. Through the power of OpenShift, users can now deploy their applications on the cloud utilizing the power of FPGAs, easier than ever.

Accelerated Machine Learning from your browser

Data scientists and ML engineers can now speedup their applications from their browser using the power of FPGAs. ML developers can now run up to 15x faster their applications on Jupyter notebooks from their browser with zero code changes. ML developers are welcome to evaluate their applications for free on the new InAccel portal. InAccel, a pioneer in making FPGAs accessible and scalable, is providing the benefits of FPGA acceleration and utilization from your browser using the familiar Jupyter platform.

The need of an Accelerator Operating System

InAccel FPGA orchestrator abstracts away the available FPGA resources serving as an OS layer and “kubernetes-alike” layer for the applications that need to be deployed on FPGAs. The main advantages that InAccel’s orchestrator offers are easy deployment, automated resource management and instant scaling

Data scientists, ML engineers and HPC users can now speedup their applications using the power of FPGAs from their browser. InAccel has made available as a demo a framework that allow any user to instantly log in in the platform and start accessing the Xilinx Vitis™ libraries on a cluster of Alveo™ cards. The framework is built on top of Jupyter Hub.

InAccel FPGA Operator: Simplifying FPGA management in Kubernetes

Helm is a package manager for Kubernetes that allows developers and operators to more easily package, configure, and deploy applications and services onto Kubernetes clusters. The InAccel FPGA Operator is a Helm chart deployed, cloud-native method to standardize and automate the deployment of all necessary components for provisioning FPGA-enabled Kubernetes systems.

FPGA integration easier than ever

InAccel has developed a unique FPGA orchestrator that allows easy deployment, instant scaling and dynamic resource management of the FPGA clusters. InAccel’s FPGA orchestrator allows the deployment of FPGAs as a Platform ready to be used by the software developers just like any other computing system. InAccel FPGA orchestrator abstracts away the available FPGA resources serving as an OS layer and “kubernetes-alike” layer for the applications that need to be deployed on FPGAs allowing easy integration with any framework.

What FPGAs can learn from the CPU and the GPU world: Automatic deployment, scaling and dynamic resource management

Accelerate ML workflows with Kubeflow using InAccel FPGA orchestrator

In this article we show how Kubeflow can be used to perform a design space exploration and hyper parameter tuning faster using Kubeflow and utilizing the power of FPGAs.

FPGA deployment and scaling on distributed systems; the missing layer

, InAccel developed world-first FPGA orchestrator that allows easy deployment, instant scaling and seamless resource management of the FPGA clusters. InAccel’s FPGA orchestrator allows the deployment of FPGAs as a Platform ready to be used by the software developers just like any other computing system. InAccel FPGA orchestrator abstracts away the available FPGA resources serving as an OS layer and “kubernetes-alike” layer for the applications that need to be deployed on FPGAs.