Today, Azure announces the general availability of Azure ND A100 v4 Cloud GPU instances—powered by NVIDIA A100 Tensor Core GPUs—achieving leadership-class supercomputing scalability in a public cloud. For demanding customers chasing the next frontier of AI and high-performance computing (HPC), scalability is the key to unlocking improved Total Cost of Solution and Time-to-Solution. 

Simply put, ND A100 v4—powered by NVIDIA A100 GPUs—is designed to let our most demanding customers scale up and scale out without slowing down.

Benchmarking with 164 ND A100 v4 virtual machines on a pre-release public supercomputing cluster yielded a High-Performance Linpack (HPL) result of 16.59 petaflops. This HPL result, delivered on public cloud infrastructure, would fall within the Top 20 of the November 2020 Top 500 list of the fastest supercomputers in the world, or top 10 in Europe, based on the region where the job was run.

Measured via HPL-AI, an artificial intelligence (AI) and machine learning (ML)-focused High-Performance Linpack variant, the same 164-VM pool achieved a 142.8 Petaflop result, placing it among the world’s Top 5 fastest known AI supercomputers as measured by the official HPL-AI benchmark list. These HPL results, utilizing only a fraction of a single public Azure cluster, rank with the most powerful dedicated, on-premises supercomputing

Over the past year, it’s become clear that businesses navigating these unprecedented times require a new level of agility. Core to this agility is achieving a level of software development excellence that was once thought unimaginable. When it comes to empowering all developers with limitless scale, choice, and possibilities, Microsoft Azure has their back.

Our commitment to developers is to make Azure the best cloud for developing intelligent applications that harness the power of data and AI. At Microsoft Build, we are announcing several exciting new capabilities and offers that make it easy and cost-effective for developers to get started with Azure data and AI services.

Innovate with Azure database services

We are announcing several new capabilities that empower developers to innovate with Azure’s database services. Azure SQL Database’s ledger capability, in preview, provides cryptographic verification for sensitive records. Customers like British Petroleum are already benefiting from this exciting feature. Azure Synapse Link for Microsoft Dataverse provides immediate insights from Microsoft Dynamics 365 and Microsoft Power Platform applications, while Microsoft Power BI streaming dataflows remove any bottleneck from signals to insights. There are also several updates to Azure Cosmos DB to help developers build and modernize high-performance applications at any scale,

Over the past year, Microsoft has teamed with Ball Aerospace to develop the prototype for the Commercially Augmented Space Inter Networked Operations (CASINO) Program Office, facilitated by the Defense Innovation Unit, demonstrating agile cloud processing capabilities in support of the United States Space Force.

The rising number of satellites proliferating in low earth orbit (LEO) presents a new data challenge for the ground segment of missions—a segment that is often overlooked. For increasingly distributed mission environments, Microsoft Azure delivers the processing power and analysis necessary for these large datasets.

The CASINO Program Office demonstrated fast, flexible, and extensible commercial capabilities for ground processing in support of defense missions. The project also confirmed the potential to transform the analysis of space data across a wide array of industries, including agriculture, ecological study, sustainability, and disaster response.

Microsoft-Ball Aerospace team demonstrates a new method of space analytics

This project represents a huge leap forward in reducing the time to actionable insight—if you are on the ground in a tactical edge vehicle or located at a command center, users can obtain necessary information accurately, quickly, and securely.

To execute the demonstrations, the team transmitted simulated overhead persistent infrared (OPIR) data through