- Get link
- X
- Other Apps

For the work of artificial intelligence, as is known, huge computational powers are needed. Now the most common components are used as electronic components (with the exception of the most powerful systems like IBM Watson). Intel solutions from consumer Core to Xeon are typically used. But all of them were created for several other tasks that do not involve working with neural networks. Therefore, the company management decided to develop a processor that will be sharpened specifically to work with AI systems.
The first specialized processor for AI from Intel was named Nervana Neural Processor (NNP), and the release of the first models is scheduled for the end of this year. The new Nervana processor is an ASIC-chip (application-specific integrated circuit), and it is primarily aimed at performing tasks in the field of deep machine learning and self-learning. The processor is based on data busses that provide bidirectional exchange of large amounts of information. Thus, it will be possible to combine several Nervana processors into one practically without loss of work efficiency. This approach allows the processor to work with multi-level neural networks. In addition, as stated by representatives of Intel,
“The main feature of the Nervana NNP processor is that it is based on mathematics with integers with a reduced Flexpoint accuracy. The advantage of its use is the increase in performance when performing tasks, reducing delays and increasing the speed of data transmission on external tires. In addition, Nervana can perform at the hardware level the functions of matrix addition, intersection finding and a number of other operations. And special algorithms allow you to dynamically change the memory of the processor so that its architecture becomes optimal for performing a specific operation. "
Despite the vast experience of Intel in the field of creating processors, when entering the market for chips for AI, it will have to face such giants as Qualcomm, IBM and NVIDIA, which have long been engaged in the production of components in this area. So watch this race will be very interesting.
The article is based on materials .
- Get link
- X
- Other Apps
Comments
Post a Comment