uploads/2019/04/A4_Semiconductors_NVDA_Go-to-market-partners-1.png

NVIDIA’s Go-to-Market Partners Make GPU Deployment Easy

By

Updated

NVIDIA’s go-to-market partners

NVIDIA (NVDA) has been strengthening its products by working alongside industry partners to identify the applications where AI (artificial intelligence) and GPU (graphics processing units) accelerated computing can make a difference. It then makes these solutions available to all industry players big and small through NGC (NVIDIA GPU Cloud) and its Deep Learning Institute.

NVIDIA expanded its go-to-market strategy to include OEMs (original equipment manufacturers), CSPs (cloud service providers), volume server partners, and DGX-ready data centers.

Article continues below advertisement

Reference architecture

At its 2019 Investor Day, NVIDIA’s executive vice president of Worldwide Field Operations, Jay Puri, explained the company designs the DGX server but a customer cannot realize its full potential without proper network and storage appliances. He stated that NVIDIA worked with storage and network firms like Cisco (CSCO) and Mellanox (MLNX) to develop pre-configured reference architectures for different workloads, which can accelerate at the data center level.

DGX-ready data centers

Jay Puri stated that another challenge in front of NVIDIA was proof of concept, which means proving to customers that NVIDIA’s solutions will bring significant performance improvement in their workloads. This whole process took almost six months, which meant data center had lengthy design win cycles. The company formed alliances with some scale-up data centers like Colovore.

Puri explained if the customer data centers did not have proper cooling and density requirements needed for scale-up computing, NVIDIA would do so in Colovore data centers and help them build out such a data center. This move reduced the proof of concept time to a few days and accelerated adoption of NVIDIA’s data center solutions.

Volume server partners and cloud companies

Many companies are doing inference on high-volume servers. Jay Puri stated that NVIDIA has partnered with volume server companies like Hewlett Packard Enterprise (HPE) and Lenovo to integrate its T4 GPU for inference in their servers. These servers are NGC-ready, which means customers can access the algorithms and models NVIDIA created for various industry domains and build from there.

NVIDIA has also partnered with cloud companies like Amazon Web Services (AMZN) to provide GPU-as-a-service, which allows small companies that do not have a data center of their own to perform deep learning and machine learning on the cloud.

NVIDIA is making its deep learning and AI solutions accessible to as many developers as possible in order to boost the adoption of its products.

Check out all the data we have added to our quote pages. Now you can get a valuation snapshot, earnings and revenue estimates, and historical data as well as dividend info. Take a look!

Advertisement

More From Market Realist