With the growth of artificial intelligence one of the technology firms that has been among the largest beneficiaries is NVIDIA Corporation. This is because graphics processing units which NVIDIA makes for gaming are also used in artificial intelligence applications as these chips are able to process huge amounts of data at extremely high speeds. Consequently NVIDIA has been dominant in the high-end processor market as its chips are widely used in training artificial intelligence systems.
In China Baidu has emerged as the technology firm with a highly advanced artificial intelligence program and for a long time has used graphical processing units developed by NVIDIA. Only recently the two firms inked an agreement that will see them partner in a variety of fields including autonomous cars and cloud computing.
Advanced Micro Devices
Baidu also recently entered into a partnership with Advanced Micro Devices, a rival of NVIDIA, which will see AMD’s graphical processing units used in Baidu’s data centers. This is the first time that AMD is seriously courting the artificial intelligence market and it has begged the question whether NVIDIA is being sidelined by Baidu.
Currently NVIDIA’s market share is in a healthy position and is growing at an exponential rate. In the last first quarters data-center revenue for NVIDIA has been growing at an average year-over-year rate of 174%. Partly this has been because NVIDIA has faced little competition and this has allowed the tech firm to become the supplier choice for most of the big data centers that are running artificial intelligence applications. Its dominance which is thanks partly to an early lead does not look like will be shaken up anytime.
AMD’s new graphical processing units known as Radeon Instinct are aimed at the high-end market and are designed with high-performance computing in mind. The chipmaker’s partnership with Baidu will thus offer AMD credibility and help establish the firm as potential competitor to NVIDIA in the rapidly growing artificial intelligence space. This comes at a time when the compounded annual growth rate for artificial intelligence chipsets is projected to be 62.9% in the next half a decade.
Tests that were conducted on DeepBench, an open-source deep-learning benchmarking tool developed by Baidu, demonstrated that the Vega graphical processing units from AMD compared favorably with the Telsa P100 graphical processing units from NVIDIA. However, NVIDIA is slated to be releasing the Telsa Volta V100 later in the year and it will be interesting to see how AMD’s chips will compare with the new release from its rival.