Intel® FPGA Acceleration Hub

Accelize empowers FPGA-Acceleration-as-a-Service, bringing the benefits of FPGA computing to the cloud services market. While FPGAs are becoming popular as cloud workloads acceleration platforms, programming FPGAs and monetizing FPGA “designs” in the cloud are challenging. The Accelize* solution, including QuickStore*, QuickPlay*, and QuickAlliance* enables a paradigm shift in how the cloud industry leverages the benefits of FPGAs.

Adaptive Microware provides the Adaptive Application Aware Accelerators (A4) Suite for Intel FPGA designs processing multiple streams of compressed video ingest, such as security, machine learning, and data analytics. H.264 / H.265 video decoding and transcoding solutions include scaling, cropping, and color space conversion offloading CPU resources and resulting in both higher video throughput performance and reduced system power.

Algo-Logic Systems builds FPGA accelerated Gateware Defined Networking* (GDN) solutions that achieve high throughput with minimal power and sub-microsecond latency. Algo-Logic Systems provides pre-built intellectual property (IP) cores and full turn-key solutions that map logic into FPGA products for three main industries—finance and high-frequency trading, data centers, and Internet of Things (IoT) solutions.

With its innovations, the b<>com Institute of Research and Technology (IRT) is taking part in the European digital transformation. Its 230 researchers develop tools, products, and services that make everyday life easier. They focus on two fields of research—hypermedia (ultra high-definition images, 3D sound, smart content, virtual, and augmented reality) and more agile ultra-high speed networks (cloud, cybersecurity, ultra high-speed mobile, network resilience, and the IoT).

Bigstream makes big data, machine learning, and artificial intelligence workloads faster through a technique called hyper-acceleration. Bigstream developed hyper-acceleration to deliver orders-of-magnitude performance gains for Apache Spark* using hardware and software accelerators. Hyper-acceleration of big data and machine learning workloads is achieved using advanced compiler technology and transparent support for FPGAs. Unlike other approaches, Bigstream requires no application code changes or special application programming interfaces (APIs).

CAST provides a broad line of IP cores and accelerator functions. Its IP cores product line includes 8 bit, 16 bit, and 32 bit processors, peripherals, buses, network interfaces, communications devices, compression, and encryption functions. CAST also offers accelerator functions for data and video compression. Established in 1993 with a focus on making IP cores practical and affordable for mainstream designers, the company has developed a reputation for high-quality products, simple licensing, and responsive technical support.

CTAccel Limited was formed in 2016 by a team of FPGA researchers and engineers, most of them had more than 10 years of FPGA development experience. By leveraging FPGA-based computing technology, CTAccel Image Processing Accelerator (CIP) with Intel PAC enables our data center customers to achieve high throughput, low and deterministic latency image processing. The image processing capabilities supported by CIP include jpeg codec, webp encode, lepton codec, resizing/crop. These functions are widely used in thumbnail generation, image transcoding and lossless image compression which are seamlessly integrated into most used applications in Data Centers.

Looking beyond parallelization and focusing on customization and specialization for energy-efficient computing, Falcon Computing provides automated compilation tools, a runtime management tool, and acceleration libraries to enable the scalable and seamless integration of heterogeneous accelerating technologies into existing data center infrastructures, with focus on FPGA-based acceleration. Falcon Computing brings orders-of-magnitude performance and power efficiency improvement for key big-data applications, including machine learning, computational genomics, and precision medicine.

iAbra delivers end-to-end enterprise neural network creation platform. Designed for embedded applications and for leveraging the efficiency and power advantages of FPGAs, the iAbra PathWorks toolkit provides all the integrated components you need to go from example data to high-performance, low-latency, and low-power inference.

Levyx is an enterprise software company that develops high-performance, low-latency data engines optimized for solid-state drives (SSDs) and advanced hardware, such as Intel® Arria® 10 FPGAs, advanced Non-Volatile Memory Express (NVMe) storage (For example, Intel Optane™ SSD and 3D XPoint™ storage technology), and multicore processors.

Real-time data analytics using FPGA acceleration as a service delivers levels of performance, low latency, and workload customization to cloud and enterprise users not possible any other way. The Intel Programmable Acceleration Card with Intel® Stratix® 10 SX FPGA provides the high-bandwidth network interfaces and deep memory needed for in-line processing of streaming data with inferencing and off-load processing of compute intensive functions supported by our Sira™ acceleration function units. Just as important, Intel® Acceleration Stack for Intel® Xeon® CPU with FPGAs integrates seamlessly with our Arka™ runtime services to expose FPGA acceleration as a service to upper-layer APIs, streamlining software development and design reuse.

Myrtle accelerates performance-critical workloads on FPGAs and has realized multiple proprietary deep learning networks as silicon designs so that they execute at a latency and power point that makes them usable in real-world situations. Myrtle has significantly accelerated the time to market for FPGA designs by automating the translation of artificial intelligence network designs using compiler technology targeting the Intel Arria 10 FPGA.

Napatech helps companies to reimagine their business by bringing hyperscale computing benefits to IT organizations of every size. We enhance open and standard virtualized servers to boost innovation and release valuable computing resources that improve services and increase revenue. Our Reconfigurable Computing Platform™ is based on a broad set of FPGA software for leading IT compute, network and security applications that are supported on a wide array of FPGA hardware designs.

rENIAC solves IO bottlenecks resulting in latency reduction and increased throughput for critical workloads such as NoSQL, Search, Graph, SQL and AI. This is supported in public cloud, hybrid and on-premise data centers without software changes to existing applications. The company’s Distributed Data Engine is architected to benefit databases, file systems, networking and storage solutions while freeing more CPU resources to creating business value.

Swarm64 enables your IT to handle large amounts of high-velocity data and eliminates the risks and costs inherent to introducing new IT systems. The Swarm64SDA uses the latest generation Intel Arria 10 FPGAs to deliver industry-leading performance.