When it comes to AI artificial intelligence and deep learning, I believe that readers and friends who pay attention to the technology circle must be familiar with it. The advent of the era of big data has provided inexhaustible data samples for the development of AI artificial intelligence technology; and deep learning is the main driving force of artificial intelligence. Gartner, a world-renowned research organization, pointed out that by 2018, 80% of data scientists will incorporate deep learning into their toolkits.
However, the success of machine learning and artificial intelligence projects, in addition to relying on data and algorithms, also requires a fusion of skills, infrastructure and business recognition.
On April 17, 2018, Hangzhou Acceleration Cloud Information Technology Co., Ltd. (referred to as Acceleration Cloud), an industry-leading manufacturer of heterogeneous acceleration and business offloading solutions, was committed to providing an overall solution for heterogeneous computing acceleration and officially launched across Beijing and Shanghai. , Chengdu and Xi'an, the "Accelerating New Technology, Driving Smart Future" technology summit and new product launch conference. Many key manufacturers and institutions in the industry, including Intel, Cytech, Tencent, JD.com, Meituan, Kingsoft, iFLYTEK, Lenovo, Sogou, Xidian, Tsinghua, Beijing Institute of Technology, Institute of Computing Technology, Chinese Academy of Sciences, etc. The organization calculates future possibilities.
At this press conference, Acceleration Cloud released four major innovative products and three solutions, and invited experts from Intel and Cytech to share solutions such as industrial Ethernet, FPGA-based OpenCL and gene acceleration with the guests at the venue. .
The four major products launched by Acceleration Cloud this time include two series of hardware acceleration products (SC-OPS, SC-VPX), two IP libraries (FDNN, FBLAS), and the three major solutions include deep learning solutions, High-performance computing and digital signal processing solutions, edge computing solutions.
Among them, SC-OPS is the world's first Intel Stratix 10 FPGA accelerator card launched by Accelerator Cloud. It adopts Intel's latest 14nm process Stratix10 GX2800 FPGA device, which can be widely used in data centers, cloud computing, machine vision, deep learning, and high-performance computing. , simulation, finance and other fields; SC-VPX is the VPX blade acceleration platform with the highest computing density in the world. It adopts Intel Stratix 10 GX2800 device and is compatible with GX1650. It constructs the industry's advanced, flexible and efficient signal processing and deep learning architecture, mainly targeting colleges and universities. Rapid construction of product prototypes and algorithm development and application in the fields of radar, communication, and deep learning in the research institute and other units.
The deep learning acceleration library FDNN is the first FPGA acceleration library in China that supports general-purpose convolutional neural networks. Based on RTL-level code, it can provide high performance and flexible configuration features; the high-performance computing acceleration library FBLAS is the industry's highest performance acceleration library. RTL level math acceleration library.
Acceleration Cloud launched a complete set of FPGA-based deep learning acceleration solutions, including SC-OPM/SC-OPF/SC-OPS acceleration cards and FDNN acceleration libraries, to meet customers' requirements for high-performance and flexible acceleration of deep learning. In order to facilitate customers to use high-level language development, Acceleration Cloud provides a complete OpenCL heterogeneous development environment based on FPGA to quickly implement user-defined deep learning acceleration solutions. At the same time, Acceleration Cloud also provides fast deep neural network customized acceleration services.
According to the requirements of digital signal processing systems such as radar and communication, combined with Intel's latest 14nm process Stratix10 FPGA series, Acceleration Cloud provides a complete set of digital signal processing solutions that combine hardware and software to achieve high-performance matrix operations (matrix operations). Multiply, transpose, inverse, QR decomposition) and ultra-fast FFT (Fourier transform). In order to facilitate customers to use high-level language development, Acceleration Cloud provides a complete OpenCL heterogeneous development environment based on FPGA to quickly implement user-defined signal processing acceleration solutions.
Accelerated cloud edge computing solutions, using high-performance Intel Arria10 GX660 devices, with modular design, strong real-time characteristics and high-performance algorithm IP acceleration, a complete OpenCL heterogeneous development environment, can realize a new generation of high-performance edge computing gateways, Used in various industrial environments.
sound:
"Artificial intelligence has entered our lives, but there are still bottlenecks in the future development, which requires breakthroughs in hardware technology and algorithms. Heterogeneous computing is the future trend of computing architecture, and FPGA is the perfect choice for realizing heterogeneous computing. Accelerating cloud innovation Heterogeneous computing acceleration platform solutions, with high performance, high efficiency, low latency, programmability and remote reconfigurability, are very suitable for the needs of elastic business on the cloud. We hope that through our technology, we can help more Many enterprises have implemented deep learning and won the first opportunity in the era of big data.”
——Wu Gang, founder and CEO of Accelerator Cloud