Why Is Qualcomm Bringing AI Capability For Cloud

Expanding its existing capabilities of integrating AI capabilities in consumer devices including smartphone, wearables, camera etc Qualcomm on Wednesday, during its AI Day event, introduced its AI and machine learning capabilities for the cloud infrastructure.

The company launched Qualcomm Cloud AI 100 platform which is built on the company’s Snapdragon technology and is claimed to be much faster than any other solution available in the industry and consumes less power. The company claimed  the new chip offers more than 10x performance per watt over the industry’s most advanced AI inference solutions deployed so far.

So, Why Qualcomm Is Moving To Cloud?

The company has already been providing AI capabilities to consumer devices like smartphones, cameras and IoT units but bringing the same to cloud was a natural evolution as the enormous amount of data that is generated from these devices is going to be analysed in the data centre.

And, it makes more sense, for Qualcomm as well as for the industry.

With 5G round the corner and IoT is fast becoming mainstream, the devices connected to the networks will send enormous amount of data to the data centres.  Similarly, more and more firms are deploying some sort of cloud solutions or the other, and the same is going to pump in unprecedented volume of data to the servers. Also, this data need to be processed, analysed and made to put for proper use. The Qualcomm Cloud AI 100 platform uses AI and machine learning to process data on the cloud and the company claims it can process trillions of data per second.

The Qualcomm Cloud AI 100 platform incorporates Snapdragon chipset on a 7nm process node. The platform supports leading software stacks, including PyTorch, Glow, TensorFlow, Keras, and ONNX. The chip is expected to be commercial y available in the second half of 2019.

Entering into cloud space also opens up a new revenue source for the San Diego based chipmaker. On its AI Day Qualcomm expanded its partnership with Microsoft and Facebook though the firm did not elaborate on the road ahead. However, it can easily be guessed that companies like Facebook that generate billions on GBs of data every minute needs high-performance machine learning-led analytics solution to filter through the data. And Qualcomm’s end-to-end solution comes in handy.

 

Also Read

Leave a Reply

 
%d bloggers like this: