Dell helps customers achieve high performance in AI and GenAI projects with enterprise storage and infrastructure innovations AI NVIDIA DGX SuperPOD.
Arthur Lewis, presidente dell’Infrastructure Solutions Group di Dell Technologies
Storage performance is critical to the success of AI and GenAI initiatives.
Customers rely on us to constantly push the boundaries of storage innovation, such as eliminating data access bottlenecks that limit the throughput and scalability of compute-intensive applications. Our response to these requests is fast, efficient and secure access to data to transform it into the potential offered by AI and GenAI.
Today, with the new software updates of PowerScale OneFS, companies can prepare, train, optimize, and make inference with AI models faster. With new PowerScale all-flash storage systems based on the latest generation of Dell servers PowerEdge you can achieve up to double the performance in read and write operations streaming.
PowerScale will offer a new smart scale-out capability to increase the performance of each compute node by improving GPU utilization resulting in increased storage throughput for training, checkpointing and inference of AI systems.
Dell PowerScale being validated on NVIDIA DGX SuperPOD
Through the collaboration between Dell and NVIDIA, customers will benefit from a validated combination of NVIDIA DGX systems, Dell PowerScale storage, and NVIDIA Quantum-2 InfiniBand and Spectrum Ethernet networking solutions to increase the speed and efficiency of AI storage. Dell’s solution is expected to be the first Ethernet storage solution validated on NVIDIA DGX SuperPOD.
With the validation of Dell PowerScale on NVIDIA DGX SuperPOD, NVIDIA’s turnkey AI data center infrastructure, customers will be able to accelerate their AI and GenAI initiatives by leveraging Dell network-attached storage systems. NVIDIA DGX SuperPOD includes the NVIDIA AI Enterprise software platform to create a secure and stable full-stack AI supercomputing solution.
Extending AI to data wherever it resides
With nearly 87% of companies pursuing multicloud strategies, Dell offers customers the freedom to process data wherever they see fit – on-premises, at the edge or in public clouds.
Dell APEX File Storage for Microsoft Azure, the latest addition to the Dell APEX Storage for Public Cloud portfolio, introduces enterprise-class performance and management capabilities for working with files within Microsoft Azure. Customers will be able to easily meet the needs of high-performance AI and machine learning applications such as Azure OpenAI Service e Azure AI Vision.
Dell APEX File Storage within AWS and Azure allows customers to leverage cloud-native AI and GenAI workflows on data residing on-premises or in public clouds by improving data access and movement. For example, through the collaboration between Dell and Databricks, customers can choose from a variety of Large Language Models (LLMs) and use Databricks libraries MosaicML to retrain a fundamental model with data residing on Dell APEX File Storage, thus gaining flexibility within environments multicloud.
Collaboration with Databricks leverages partnerships across Dell’s data management ecosystem to help customers accelerate their AI and analytics initiatives. And with Dell’s modern, open data lakehouse, customers can gain insights from data wherever it resides through analytics software integration Starburst con i server Dell PowerEdge and with Dell PowerScale, Dell ECS, and Dell ObjectScale file and object storage systems.