Panasas, Inc. unveiled its portfolio storage solutions for the complex data challenges.
Two ActiveStor systems will accelerate advanced AI/ML training, modelling and simulation, high performance data analytics, and massive-scale data workloads.
All-NVMe ActiveStor Flash and massive-capacity ActiveStor Ultra XL join the company’s price/performance storage platform, ActiveStor Ultra. This portfolio is powered by the firm’s PanFS data engine parallel file system for Linux clusters.
Mark Nossokoff, research director and lead storage analyst, Hyperion research, remarked: “We’ve witnessed a significant 13% increase in on-premises HPC spending from 2020 to 2021 and are projecting a market growth CAGR of 6.4% over the 2026 forecast horizon. The growth is largely driven by organisations’ emerging use cases, such as AI/ML and big data analytics, that require new technologies and solutions from the HPC space to address some of today’s biggest challenges. We’re excited to see an HPC storage veteran vendor like Panasas embrace these newer workloads with their new portfolio of ActiveStor storage appliances.”
The company’s PanFS data engine is a parallel file system and the heart of the ActiveStor storage family. It is simple to manage, yet it delivers performance, scalability, and reliability. The PanFS software is a reliable and autonomic data management engine built on 20+ years of hardening through commercial deployment. This data engine takes the complexity out of high performance data environments and provides high-speed direct file-based data access to applications with no manual administration required. It orchestrates all data management to automatically recover from failures, continuously and balance data loads, and scrub and encrypt stored data for high level of protection.
The introduction of the ActiveStor portfolio is validation of the PanFS data engine’s robustness, as well as the company’ ability to deploy tailored data solutions that meet customers’ specific workload, footprint, and environmental requirements.
“Forward-thinking organisations are harnessing the exponential data growth we’re seeing today to build a brighter tomorrow, from developing life-saving pharmaceuticals to designing energy-saving autonomous vehicles. The technology they use to store, deliver, and secure their data is vital to their success,” said Tom Shea, president and CEO. “Choosing the optimal data foundation for your organisation’s specific needs and goals is not just a good business strategy – it’s imperative. That’s why we’re introducing a portfolio of products for the first time in our history. Our storage systems have been purpose-built to provide a diverse range of high performance data solutions for all workload types at the best total cost of ownership.”
ActiveStor products are suited to a variety of computing environments across industries.
Activestor Flash ASF 100
ActiveStor Flash. This all-NVMe ActiveStor Flash provides scratch and small – and random-file storage performance. It is for AI/ML training, trading strategy backtesting, as well as life sciences and electronic design automation projects.
ActiveStor Ultra XL
ActiveStor Ultra XL. This high-capacity ActiveStor Ultra is for massive-scale data environments in addition to workloads with cooler and reference datasets and large file sizes, such as seismic resource exploration; scientific, academic, and government research; manufacturing; and M&E.
ActiveStor Ultra. This hybrid ActiveStor Ultra has been a stalwart high performance mixed workload solution for traditional and enterprise HPC. It delivers performance for high performance data analytics, complex modeling and simulation, molecular imaging techniques in life sciences, and converged HPC/AI data use cases.
The company was at World Summit AI Americas in Montréal, Canada on May 4–5, 2022, and will be at ISC High Performance in Hamburg, Germany on May 29–June 2, 2022.