Traditional approaches to AI infrastructure often result in silos of servers and storage that starve AI workloads or are over-spent on capacity. These common complexities of legacy solutions can hinder enterprises from moving into the new era of AI.
This is why Pure Storage and NVIDIA developed AIRI®—the industry’s first complete full-stack AI-ready infrastructure. AIRI is tuned from software to hardware to eliminate bottlenecks and enable data scientists to get up and running in a few hours, not weeks or months. Without the limitations of legacy infrastructure and complexities, AIRI allows modern innovators to keep their focus on algorithms instead of infrastructure, while giving organizations a competitive edge.
Since AI workloads process massive amounts of data from unstructured sources, they require a solution such as AIRI, powered by Pure’s unified fast file and object (UFFO) storage platform FlashBlade®. AIRI offers a simple, flexible and massively parallel infrastructure to meet demands for any AI initiative, at any scale. AIRI flexibly adapts to business demand as AI models move from prototyping to deployment, with one universal AI system that offers right-sized resources for every workload, and consolidates silos into a single, elastic AI infrastructure.
The NVIDIA® DGX A100 systems and FlashBlade platform within AIRI replace the performance of racks of legacy technology, offering data scientists the power of a large supercomputer for any AI initiative. AIRI offers graphics processing units (GPUs), scale-out flash, and remote direct memory access (RDMA) fabric that move large data loads.
A Leader in AI Innovation
Pure was recently selected as the winner of the “Best AI Solution for Big Data” award in the fourth annual AI Breakthrough Awards program for our AIRI infrastructure solution. AI Breakthrough is a leading market intelligence organization that recognizes the top companies, technologies and products in the global Artificial Intelligence (AI) market.
The mission of the AI Breakthrough Awards is to honor excellence and recognize the innovation, hard work and success in a range of AI and machine learning related categories, including AI platforms, Deep Learning, Smart Robotics, Business Intelligence, Natural Language Processing, industry specific AI applications, and many more. This year’s program attracted more than 2,850 nominations from over 17 different countries throughout the world.
AIRI from Pure Storage offers new, modern ‘breakthrough’ technologies that help enterprises tap the power of AI for faster innovation and a competitive edge.
James Johnson, managing director of AI Breakthrough, says: “The unfortunate reality of too many of today’s AI initiatives is that they are stalled by the complexity and rigidity of legacy technology, as well as the intricacies of a ‘do-it-yourself’ approach. AI requires a completely new infrastructure to truly facilitate faster innovation and obtain actionable, modern analytics. AIRI from Pure Storage offers new, modern ‘breakthrough’ technologies that help enterprises tap the power of AI for faster innovation and a competitive edge.”
Driving Customer Breakthroughs
The largest medical center in Taiwan, Chang Gung Memorial Hospital, knows the importance of AI to the advancement of medical care. They brought in Pure and AIRI to improve efficiency and accelerate genetic research. Since the introduction of FlashBlade, system learning time has been reduced from seven days to just one day, and computing efficiency has increased seven times. With Pure as their infrastructure choice, Chang Gung can keep improving data security, stability, and computing speeds to drive medical breakthroughs and enhance doctor-patient relationships.
Core Scientific is a leader in high-performance infrastructure and software solutions, powering many of the world’s blockchain companies. They collect billions of data points every day, and were experiencing sub-par performance on their applications. Since implementing AIRI, they have seen an eight times increase in the volume of data they are able to process. This enables their data scientists to analyze massive amounts of data on a scale, not previously possible with legacy IT infrastructure.