The most recent version of PowerScale allows for improved searchability, better density, faster networking, and AI-ready performance.
AI is now the foundation of companies looking to compete in the data-driven economy of today, not just a far-off vision of the future. Businesses are hurrying to employ AI to automate complex processes and innovate customer interactions. Dell Technologies understands that AI’s revolutionary potential depends on both cutting-edge algorithms and infrastructure.
They are pleased to present the most recent enhancements to PowerScale, it unstructured data storage platform, which is intended to assist companies of all sizes in utilising AI to discover new opportunities. With this introduction, Dell offers increased searchability with the new metadataIQ technology, quicker networking with 200Gb Ethernet and InfiniBand choices, and higher storage density with 61TB SSD drives and 24TB HDDs.
Here is a closer look at the features of the most recent OneFS version and its implications for AI-enabled businesses going forward:
Faster Networking for Accelerated AI Workflows
The speed at which a system can access and transfer data at scale is just as important to the efficacy of AI processes as its overall power. PowerScale offers 200GbE and 200Gb InfiniBand connectivity to remove bottlenecks and advance AI performance. These improvements allow data to flow smoothly across all stages of the AI pipeline, with a 220% streaming write and 99% streaming read jump over the previous year.
What does this mean for AI workloads?
AI Pipeline Optimisation: Ultra-fast, low-latency connectivity guarantees that even large datasets move seamlessly between storage and computation, from data import to model training and inference. Businesses may now conduct more frequent training checkpoints and expedite model training for large-scale AI implementations, such as Generative AI applications, by reducing data bottlenecks.
Improved GPU Performance: By simplifying data delivery, faster networking has a direct effect on GPU performance for enterprises implementing GenAI environments. Businesses can gain more insights and make data-driven choices more quickly by using PowerScale to handle larger datasets in less time.
Higher Storage Density for Efficiency and Cost Savings
Handling enormous datasets while controlling expenses is necessary to be competitive in AI, a problem that contemporary firms are all too familiar with. With the all-flash PowerScale F710, a density milestone that gives companies significant advantages, PowerScale now offers 61TB SSD drives. In order to satisfy TCO-optimized client expectations for demanding hybrid workloads, PowerScale has recently expanded its product line to include 24TB HDDs for the A300, A3000, H700, and H7000 nodes.
Key advantages include
Lower Data Centre Costs: Energy usage and colocation costs can be significantly reduced by up to 50% when storage footprints are reduced. It’s a green effort that supports sustainability and corporate efficiency objectives.
Developing Models for Generative AI: Larger Large Language Model (LLM), such as those used in Generative AI, such as NVIDIA DGX SuperPOD, are supported by the higher density. These models enable organizations to deploy AI with fewer nodes, resulting in better insights and more cost-effective implementations. Higher output and shorter latency are the direct outcomes of the 24TB drive’s superior SSD cache to HDD ratio when compared to 20TB drives for HDD-based devices.
Streamlined Data Management with Enhanced Searchability
AI is driven by data, but managing and deriving valuable insights from the vast amount of unstructured data can be difficult. The new MetadataIQ features from PowerScale transform how businesses manage metadata by revealing important context across dispersed datasets.
PowerScale enables enterprises to effectively find their data and get insights from petabytes of data and billions of files by exporting metadata into Elasticsearch databases.
Among the transformative capacities are:
Better Data Accessibility: Businesses can easily find important information concealed in unstructured sources like support databases, knowledge bases, and PDFs. More informed, real-time departmental decision-making results from faster access.
Effective AI Training: Training datasets are curated, consistent, and contextualized with the capacity to access rich metadata across dispersed clusters. By removing the need for treewalks and processing only updated or new content, this method enables organisations to maintain the efficiency of their RAG workflows while optimising compute utilisation to save time, money, and derive accuracy in Generative AI environments.
Scale to new AI horizons with PowerScale are you ready?
AI is still developing with data, and optimising its potential requires the correct storage mechanism. Dell Technologies enables companies to transform data into action, insights, and creativity through PowerScale‘s quicker networking, increased storage density, and improved searchability. These solutions enable businesses to concentrate on results rather than infrastructure, whether they are powering multibillion-dollar media projects or operating vital financial models.
More than just state-of-the-art technology, Dell PowerScale provides performance, scalability, flexibility, and simplicity, serving as the cornerstone for innovation in companies worldwide. PowerScale is the engine powering robust, future-ready AI implementations, from facilitating smooth transitions between on-premises and cloud settings to scaling massive datasets with unparalleled flexibility.
