Nvidia is paving the way for a technology, driven by deep leaning techniques that could take the search capabilities of today’s systems to a whole new level.
Nvidia has announced the launch of its new processors called Nvidia Tesla M40 and M4 graphic processing units that make it easier to search for information and is intelligent enough to spot the colour tone and emotional state of the user.
Speaking about the launch Ian Buck, vice president of the accelerated computing at Nvidia, stated that both Nvidia Tesla M40 and M4 are server-based. Both the units will have to be housed in the same sever which stores the videos as well.
The graphic product developer, which holds a monopoly when it comes to graphic card for PC’s is also measuring the way deep learning can be used to increase the capabilities of the present supercomputers. The Tesla M40 and M4 are perform different functions in a machine learning system. While the M40 is about training of the neural network in a deep learning system, the M4 deals with dense clusters of systems that put to effect the machine learning tasks.
Machine learning to help add information attached to videos could improve the accuracy of search results. Machine-learning processors could also power software models designed to analyse videos and images,
The Tesla M40 and M4 are also Nvidia’s first of a kind GPUs that are going to be used for Web hosting and serving.
The deep learning systems developed by Nvidia is also utilised in vehicles especially self-driving cars that rely on artificial intelligence to guide them through unpredictable traffic. The AI in turn relies on the deep learning technology that allows the AI to learn new information regularly. Nvidia’s flagship deep learning system for vehicles is the Tegra X1, which has been used in self-driving cars. The Tegra X1 gives the car better ability to detect in coming traffic, sign posts and objects.
Specifications wise the Tesla M40 has 3072 cores and can deliver an optimum performance of 7 teraflops. It also boasts of 5GB of GDDR5 memory and draws in about 250 watts of power to run its operations.
The M4 on the other hand can be considered to be smaller version of the M40. It is designed more for the hyperscale environments, and has been provided with 4GB of GDDR5 memory. It also draws in less energy compared to the M40 at a meagre 50 to 75 watts per hour and can deliver 2.2 teraflops of performance.