This post was originally published on this siteAt Pure//Accelerate 2018 Cody Hosterman and I delivered a session on “Moving Data Between Cloud and On-Premises Virtualized Environments“....
If you have been following Google (well… who does not?) lately, it is clear that they are going all in on Artificial Intelligence (AI). Their annual Google I/O Developer Conference just two weeks ago was peppered with major announcements and cool demonstrations. Just watch this impressive demo of Google Duplex having a real conversation with actual people, even dropping a super casual “mmhmm” into the conversation.
But it’s not just Google. Over the past few years, visionary companies such as Facebook, Microsoft, Amazon, and Tesla have all pushed the limits of AI, integrating it naturally into our lives more seamlessly than ever before.
On the other end of the spectrum, however, many enterprises are still finding it a challenge to cross the AI chasm and to get past their visionary ideas into meaningful products.
In his eponymous book, Moore outlines that the primary challenge of crossing the chasm is that the early majority are pragmatists and define if the disruptive technology will die in the chasm or cross over to the mainstream market. Simply put, what works for visionaries and innovators, does not work for the pragmatists. Pragmatists need solutions that are proven, tested and mature so that they can focus on how the technology helps their business problem and not the technology itself.
As the adoption of AI increases, Moore’s analogy is now relevant more than ever. Enterprises face major challenges with many of their AI-initiatives failing to materialize beyond some initial rudimentary success. They may lose months of productivity trying to leverage the existing, complex DIY platforms and go through the painful cycle of integrating, testing and continuously maintaining both the hardware and ever-evolving software. Due to this, they often fail with their AI initiatives even before shipping their first intelligent product or service. Businesses need an innovative solution that enables them to quickly deploy an AI platform that is supported by the industry leaders so that they can focus on building their competitive edge with AI.
Working closely with the AI visionaries such as Paige.AI, we at Pure, along with NVIDIA have incorporated practical experience and best practices and built AIRI™. AIRI is a revolutionary end-to-end AI-at-scale infrastructure to address real-world challenges that enterprises face. AIRI offers a unique capability for enterprises to cross the AI chasm with ease and benefit from the work done by these trailblazers. It is backed by enterprise-grade support for the AIRI Technology Stack that includes NVIDIA GPU Cloud Deep Learning Stack, Pure1® cloud-based management, and AIRI’s multi-node scaling toolkit.
However, even within the early adopters, enterprises are at different phases within the AI initiative. Some are well ahead in the curve using AIRI’s multi-DGX training out-of-box, while some others are in the early stages of adoption and are constrained by a limited budget. There is a significant gap in the AI infrastructure between a single DGX-1 server that is suitable for a data scientist with a limited dataset and the scaled-out infrastructure critical for large data scientist team training on a treasure trove of big data.
To address this gap, we are excited today to bring you two groundbreaking announcements at //ACCELERATE – AI is now indeed within reach of every enterprise – small or large.
First, we are announcing AIRI™ “mini” packed with two NVIDIA DGX-1 servers and Pure FlashBlade™ storage, and it delivers the performance of 25 racks of legacy infrastructure. AIRI “mini” is an affordable AI infrastructure, shattering the barrier to adoption for organizations looking to explore AI-at-scale. It is ideal for organizations growing from a single data scientist to a team of data scientists. Built on the same software and hardware management as AIRI means it is consistent, efficient and simple for IT generalists, enabling enterprises to start with mini & scale to many!
Secondly, we are building on the strong partnership with Cisco and extending the success of FlashStack™ to bring AI to more enterprises. So at //Accelerate, we are announcing a new AIRI platform with Cisco networking solutions. AIRI now enables Cisco customers to seamlessly incorporate AI-strategy and AIRI in their organization leveraging their existing fabrics infrastructure. This AIRI solution is built with Cisco’s Nexus 9300 series 36-port 100GE switch but supports other Cisco cloud-scale switches as well.
AIRI represents a giant leap to cross the chasm, empowering you to focus on what data scientists really want to do – harness the force of AI without worrying about the myriad infrastructure complexity. AIRI is here, customer-proven and shipping now. No matter where you are in your AI journey, it lets you tap the power of AI-at-scale.
To learn more, talk AIRI to us at //Accelerate or visit us at www.purestorage.com/airi
For more information on announcements this year at Pure//Accelerate, check out our overall launch blog, Data Centric Architecture Powers Digital Business.