Flexential has announced that it is the first data center globally to host ONTAP AI Test Drive, a fully optimized and tested infrastructure solution for AI workloads.
The platform, currently available in one of Flexential's Portland, Oregon, (Hillsboro) data centers, will be offered in more of the 21 markets served by Flexential in the future. For a limited time, providers are offering a test drive to allow select customers to try ONTAP AI before making an investment.
ONTAP AI testing using ImageNet data with a NetApp AFF A800 system and NVIDIA DGX-1 servers in a 1:4 storage-to- compute configuration achieved training throughput of 23,000 training images per second (TIPS) and inference throughput of 60,000 TIPS. With this configuration, clients can expect to get over 2GBps of sustained throughput (5GBps peak) with well under 1ms of latency while the GPUs operate at over 95 percent utilization.
A single AFF A800 system supports throughput of 25GBps for sequential reads and 1 million IOPS for small random reads at under 500-microsecond latencies for NAS workloads. These results demonstrate available performance headroom that can support many more DGX-1 servers as requirements increase.
ONTAP AI allows users to start small and grow as needed, and subsequently adding compute, storage, and networking to clustered con gurations without disrupting ongoing operations. Start with a 1:1 storage-to-compute configuration and scale out as your data grows to a 1:5 configuration and beyond.
NetApp’s rack-scale architecture allows organizations to start with an AFF A220 and grow as needed to scale from hundreds of terabytes to tens of petabytes with all-flash. With NetApp ONTAP FlexGroup, up to 20 petabytes of single namespace can handle more than 400 billion les.
ONTAP AI leverages NetApp Data Fabric to unify data management across the pipeline with a single platform. Use the same tools to securely control and protect data in flight, in use, or at rest and meet compliance requirements with confidence. If an issue arises in the deep learning environment, users can rely on their single point of contact and support model to help troubleshoot and provide guidance.
"Artificial intelligence and machine learning will soon be used in almost every industry, meaning that now is the time for businesses to plan how they will manage, support and optimize the technology to run these mission-critical applications," said Jason Carolan, chief innovation officer, Flexential. "Our data centers are ready to provide the power, cooling, resiliency, security, and connectivity to support these high-density AI workloads. Plus, through our professional services practice, we are able to help customers seamlessly migrate data, integrate to hyper-scale or private cloud, and provide services to secure their platforms."
In January, Flexential first announced its collaboration with NVIDIA to bring DGX-Ready data center colocation services to customers. This capability ensures Flexential data centers can support the high-density power and cooling necessary for advanced AI workloads. ONTAP AI, available in the Flexential Portland data center, allows select customers to pilot their AI workloads.
No comments:
Post a Comment