IBM Enhances Next-Gen SDS ESS 3500 to Address AI Adoption Challenges and Performance Barriers  

IBM Corp. enhenced its next-gen SDS, ESS 3500.

The product is designed to accelerate data delivery for AI workloads, be optimized for GPU compute environments, and help speed time to market with cloud-scale performance and capacity. 

In today’s hybrid cloud reality, distributed data holds the key to unlocking value through new business and operational insights. However, according to the Global AI Adoption Index 2021, conducted by Morning Consult on behalf of IBM, increasing data complexity and data silos is a top 3 barrier to AI adoption. This can be critical for transforming business processes, from supply chains and asset management to analytics. In addition, clients must also consider new challenges associated with unstructured data including navigation of the digitized world, pace of innovation, cyber-attack complexity and data sharing requirements.

 The ESS 3500 is designed to address today’s most-vexing enterprise data issues and optimized for AI- accelerated computing solutions, such as Nvidia DGX systems with GPUDirect support. Based on a client result, the company can improve AI training time as much as 70% using Spectrum Scale and Elastic Storage systems ⁽¹⁾. The solution is designed for compute-intensive workloads with the ability to scale from 46TB to ~1PB effective capacity using LZ4 compression with a 2.5x compression rate ^[3] and is projected to support over 1.8TB/s in a 20-node rack configuration ^[4]. 

 Applications and architectures that could benefit include those that previously required the use of proprietary and costly storage solutions: BC; design and simulations, especially AI and ML; stateful container environments, imaging and video content management, delivery, and streaming systems; HPC environments for research and analytics; scalable medical image archive and patient record management for healthcare; and securely accessible storage with built-in data integrity and compliance for regulated industries.

“The growing adoption of AI and Kubernetes by enterprises requires a new model that simplifies data access and scales readily as these projects grow,” said Scott Baker, CMO, IBM Storage. “IBM understands distributed file and object workload needs and we solve for myriad use cases. The ESS 3500 builds on our leadership in this market with the latest in hardware advances to help address rapidly evolving market needs, such as next generation of AI and container-based workloads.”

“Our customers have been using the capabilities of Spectrum Scale and the ESS family. They are drawn to the value of data resiliency and security with Spectrum Scale and ESS from accidental and malicious attacks that can result in loss of data,” said Michael Sedlmayer, president and supercomputer architect, Re-Store LLC. “Our customers appreciate the power, and the way IBM handles hard to solve problems better than anyone else. We are currently transitioning our sales efforts to the new ESS 3500 to reach new customers, as we have found nothing compares to the functionality and performance of the ESS with Spectrum Scale.”

In addition to AI use cases, the company’s clients have a growing portfolio of AI use cases that are transforming customer interactions, workflows, and business insights. In order to speed AI projects and adoption across the organization we think that clients need scalable solutions that can provide flexibility and growth.

“AI workloads demand powerful infrastructure that delivers cost-effective performance and scale, which is why customers tackling the most challenging AI opportunities depend on Nvidia DGX systems and Nvidia DGX SuperPOD,” said Charlie Boyle, VP, DGX systems, Nvidia Corp. “Building on Nvidia’s long collaboration with IBM, the new ESS 3500 storage system enables DGX customers to quickly and easily scale their infrastructure to speed AI-powered insights from their data.”

The ESS 3500 will be generally available on May 20, 2022. 

Ressource:
David Wohlford’s blog, marketing manager, WW cloud storage portfolio, IBM

^[1] Seagate’s Rethink Data Report 2020    
^[2] Based on data from an actual customer who used Spectrum Scale software and ESS  as compared to previous configurations using similar workloads. Disclaimer: Results based on one customer.
^[3] Disclaimer: Effective capacity based on internal testing of compression technology of various workload types and is based on publicly available documents. Actual customer capacity may vary up or down depending on type of data that is stored.
^[4] Disclaimer: Spectrum Scale can ‘allow concurrent read and writes from multiple nodes.’ Reference and performance results for 20 nodes assume workloads can be spread evenly across all nodes in the configuration.