Ubiquitous Nvidia Powers Dell EMC AI Stack 

shutterstock_255929173

Wherever there’s AI, Nvidia’s there. Wherever there’s a car trying to drive itself, Nvidia’s there. Wherever there’s a robot stacking boxes coming off a warehouse conveyer belt, Nvidia’s there. Wherever there's a data scientist on Wall Street developing a machine learning stock bet sizing application, Nvidia’s there. Wherever there’s a life sciences start-up modeling the spread of infection in hospitals, Nvidia’s there. And wherever there's a server or storage vendor building AI into its product portfolio, Nvidia’s there, too.**

With its GPU chips ubiquitous in the AI product strategies of a myriad of companies, Nvidia can watch, and lend a hand as need be, while its industry partners beat their brains out cobbling AI solutions for the broader, non-1 percent market. Some of those products will hit, more will fizzle. Either way, it’s a win for Nvidia. And as the hits pile up and AI expands, Nvidia will expand, too.

Good business model, that, Nvidia.

Last week, we reported on NetApp’s new AI platform and data-scientist-in-a-box powered by Nvidia’s DGX supercomputers. (We also reported on a new study showing that most AI projects fail, hence the rising market demand for solutions that ease AI and make failure less likely.)

Today, Dell EMC announced Ready Solutions for AI, a technology stack intended to simplify AI and relieve organizations the drudgery of sourcing and piecing together their own solutions. Instead, Dell EMC offers a set of technologies – including AI frameworks and libraries – with compute, networking and storage, along with Dell EMC consulting and deployment services. Built for deep learning workloads, Ready Solutions for AI was co-engineered by Dell EMC and Nvidia and built around Dell EMC PowerEdge servers.

Features include:

• Dell EMC PowerEdge R740xd and C4140 servers with four NVIDIA Tesla V100‑SXM2 Tensor Core GPUs. In its announcement, Dell EMC said that with 640 tensor cores, the Tesla V100 was the first, according to Nvidia, to break the 100 teraFLOPS barrier for deep learning performance
• Dell EMC Isilon F800 All-Flash Scale-out NAS storage for deep learning for analyzing large datasets concurrently

Ready Solution for AI also includes Machine Learning with Hadoop, built in partnership with Cloudera and Intel. It has a solution stack with data science and framework optimization designed to accelerate initial implementations and allow expansion of existing Hadoop environments for machine learning. Features include the Cloudera Data Science Workbench for “self-service data science,” the Apache Spark distributed data analytics engine and Dell EMC’s Data Science Provisioning Engine, which provides pre-configured containers giving data scientists access to the Intel BigDL distributed deep learning library on the Spark framework. The provisioning portal has a GUI for self-service access to hardware resources and a set of AI libraries and frameworks, such as Caffe and TensorFlow, reducing the steps required to configure a data scientist’s workspace to five clicks, according to Dell EMC.

“Organizations spend precious and costly time gathering, configuring, testing and troubleshooting machine learning environments for data scientists, at the expense of time delivering business insights,” said Bill Wagner, CEO of Bright Computing. “Dell EMC Ready Solutions for AI, with Bright Cluster for Data Science, makes it easier to get machine learning environments up and running in a cluster-ready environment to automatically scale as demand for deep learning capacity grows.”

Dell EMC cited a report (commissioned by the company) from industry watcher Forrester Research concluding that, taken together, Ready Solution for AI improves overall data science productivity up to 30 percent and reduces time-to-operations by 6-12 months, compared to do-it-yourself approaches to AI.

“AI is being driven by leaps in GPU computing power that defy the slowdown in Moore’s Law,” said Ian Buck, VP/GM, Accelerated Computing Group, NVIDIA. “Dell EMC Ready Solutions for AI with Tensor Core GPUs empower AI developers to tackle some of the greatest challenges of our time.”

** With apologies to John Steinbeck.