This week we have now launched a wave of purpose-built datacenters and infrastructure investments we’re making all over the world to help the worldwide adoption of cutting-edge AI workloads and cloud providers.
In the present day in Wisconsin we launched Fairwater, our latest US AI datacenter, the biggest and most subtle AI manufacturing facility we’ve constructed but. Along with our Fairwater datacenter in Wisconsin, we even have a number of an identical Fairwater datacenters underneath building in different areas throughout the US.
In Narvik, Norway, Microsoft introduced plans with nScale and Aker JV to develop a brand new hyperscale AI datacenter.
In Loughton, UK, we introduced a partnership with nScale to construct the UK’s largest supercomputer to help providers within the UK.
These AI datacenters are vital capital tasks, representing tens of billions of {dollars} of investments and lots of of hundreds of cutting-edge AI chips, and can seamlessly join with our world Microsoft Cloud of over 400 datacenters in 70 areas all over the world. By way of innovation that may allow us to hyperlink these AI datacenters in a distributed community, we multiply the effectivity and compute in an exponential option to additional democratize entry to AI providers globally.
So what’s an AI datacenter?
The AI datacenter: the brand new manufacturing facility of the AI period
An AI datacenter is a novel, purpose-built facility designed particularly for AI coaching in addition to working large-scale synthetic intelligence fashions and functions. Microsoft’s AI datacenters energy OpenAI, Microsoft AI, our Copilot capabilities and lots of extra main AI workloads.
The brand new Fairwater AI datacenter in Wisconsin stands as a exceptional feat of engineering, overlaying 315 acres and housing three large buildings with a mixed 1.2 million sq. ft underneath roofs. Setting up this facility required 46.6 miles of deep basis piles, 26.5 million kilos of structural metal, 120 miles of medium-voltage underground cable and 72.6 miles of mechanical piping.
In contrast to typical cloud datacenters, that are optimized to run many smaller, unbiased workloads resembling internet hosting web sites, e mail or enterprise functions, this datacenter is constructed to work as one large AI supercomputer utilizing a single flat networking interconnecting lots of of hundreds of the newest NVIDIA GPUs. In actual fact, it is going to ship 10X the efficiency of the world’s quickest supercomputer as we speak, enabling AI coaching and inference workloads at a degree by no means earlier than seen.
The position of our AI datacenters – powering frontier AI
Efficient AI fashions depend on hundreds of computer systems working collectively, powered by GPUs, or specialised AI accelerators, to course of large concurrent mathematical computations. They’re interconnected with extraordinarily quick networks to allow them to share outcomes immediately, and all of that is supported by huge storage programs that maintain the info (like textual content, photographs or video) damaged down into tokens, the small models of data the AI learns from. The objective is to maintain these chips busy on a regular basis, as a result of if the info or the community can’t sustain, every little thing slows down.
The AI coaching itself is a cycle: the AI processes tokens in sequence, makes predictions concerning the subsequent one, checks them towards the suitable solutions and adjusts itself. This repeats trillions of occasions till the system will get higher at no matter it’s being skilled to do. Consider it like knowledgeable soccer workforce’s observe. Every GPU is a participant working a drill, the tokens are the performs being executed step-by-step, and the community is the teaching employees, shouting directions and protecting everybody in sync. The workforce repeats performs time and again, correcting errors till they’ll execute them completely. By the tip, the AI mannequin, just like the workforce, has mastered its technique and is able to carry out underneath actual sport situations.
AI infrastructure at frontier scale
Goal-built infrastructure is essential to having the ability to energy AI effectively. To compute the token math at this trillion-parameter scale of main AI fashions, the core of the AI datacenter is made up of devoted AI accelerators (resembling GPUs) mounted on server boards alongside CPUs, reminiscence and storage. A single server hosts a number of GPU accelerators, related for high-bandwidth communication. These servers are then put in right into a rack, with top-of-rack (ToR) switches offering low-latency networking between them. Each rack within the datacenter is interconnected, making a tightly coupled cluster. From the skin, this structure seems to be like many unbiased servers, however at scale it capabilities as a single supercomputer the place lots of of hundreds of accelerators can practice a single mannequin in parallel.
This datacenter runs a single, large cluster of interconnected NVIDIA GB200 servers and thousands and thousands of compute cores and exabytes of storage, all engineered for probably the most demanding AI workloads. Azure was the primary cloud supplier to carry on-line the NVIDIA GB200 server, rack and full datacenter clusters. Every rack packs 72 NVIDIA Blackwell GPUs, tied collectively in a single NVLink area that delivers 1.8 terabytes of GPU-to-GPU bandwidth and offers each GPU entry to 14 terabytes of pooled reminiscence. Quite than behaving like dozens of separate chips, the rack operates as a single, big accelerator, able to processing an astonishing 865,000 tokens per second, the very best throughput of any cloud platform out there as we speak. The Norway and UK AI datacenters will use related clusters, and make the most of NVIDIAs subsequent AI chip design (GB300) which affords much more pooled reminiscence per rack.
The problem in establishing supercomputing scale, significantly as AI coaching necessities proceed to require breakthrough scales of computing, is getting the networking topology good. To make sure low latency communication throughout a number of layers in a cloud surroundings, Microsoft wanted to increase efficiency past a single rack. For the newest NVIDIA GB200 and GB300 deployments globally, on the rack degree these GPUs talk over NVLink and NVSwitch at terabytes per second, collapsing reminiscence and bandwidth obstacles. Then to attach throughout a number of racks right into a pod, Azure makes use of each InfiniBand and Ethernet materials that ship 800 Gbps, in a full fats tree non-blocking structure to make sure that each GPU can discuss to each different GPU at full line price with out congestion. And throughout the datacenter, a number of pods of racks are interconnected to cut back hop counts and allow tens of hundreds of GPUs to operate as one global-scale supercomputer.
When specified by a conventional datacenter hallway, bodily distance between racks introduces latency into the system. To handle this, the racks within the Wisconsin AI datacenter are specified by a two-story datacenter configuration, so along with racks networked to adjoining racks, they’re networked to extra racks above or beneath them.
This layered method units Azure aside. Microsoft Azure was not simply the primary cloud to carry GB200 on-line at rack and datacenter scale; we’re doing it at large scale with prospects as we speak. By co-engineering the complete stack with one of the best from our trade companions coupled with our personal purpose-built programs, Microsoft has constructed probably the most highly effective, tightly coupled AI supercomputer on the earth, purpose-built for frontier fashions.
Addressing the environmental impression: closed loop liquid cooling at facility scale
Conventional air cooling can’t deal with the density of contemporary AI {hardware}. Our datacenters use superior liquid cooling programs — built-in pipes flow into chilly liquid straight into servers, extracting warmth effectively. The closed-loop recirculation ensures zero water waste, with water solely wanted to replenish as soon as after which it’s frequently reused.
By designing purpose-built AI datacenters, we have been capable of construct liquid cooling infrastructure into the ability on to get us extra rack-density within the datacenter. Fairwater is supported by the second largest water-cooled chiller plant on the planet and can repeatedly flow into water in its closed loop cooling system. The new water is then piped out to the cooling “fins” on either side of the datacenter, the place 172 20-foot followers chill and recirculate the water again to the datacenter. This method retains the AI datacenter working effectively, even at peak masses.
Over 90% of our datacenter capability makes use of this method, requiring water solely as soon as throughout building and frequently reusing it with no evaporation losses. The remaining 10% of conventional servers use outside air for cooling, switching to water solely through the hottest days, a design that dramatically reduces water utilization in comparison with conventional datacenters.
We’re additionally utilizing liquid cooling to help AI workloads in a lot of our present datacenters; this liquid cooling is achieved with Warmth Exchanger Models (HXUs) that additionally function with zero-operational water use.
Storage and compute: Constructed for AI velocity
Trendy datacenters can include exabytes of storage and thousands and thousands of CPU compute cores. To help the AI infrastructure cluster, a wholly separate datacenter infrastructure is required to retailer and course of the info used and generated by the AI cluster. To present you an instance of the size — the Wisconsin AI datacenter’s storage programs are 5 soccer fields in size!
We reengineered Azure storage for probably the most demanding AI workloads, throughout these large datacenter deployments for true supercomputing scale. Every Azure Blob Storage account can maintain over 2 million learn/write transactions per second, and with thousands and thousands of accounts out there, we are able to elastically scale to fulfill nearly any knowledge requirement.
Behind this functionality is a essentially rearchitected storage basis that aggregates capability and bandwidth throughout hundreds of storage nodes and lots of of hundreds of drives. This allows scale to exabyte scale storage, eliminating the necessity for guide sharding and simplifying operations for even the biggest AI and analytics workloads.
Key improvements resembling BlobFuse2 ship high-throughput, low-latency entry for GPU node-local coaching, guaranteeing that compute sources are by no means idle and that large AI coaching datasets are all the time out there when wanted. Multiprotocol help permits seamless integration with numerous knowledge pipelines, whereas deep integration with analytics engines and AI instruments accelerates knowledge preparation and deployment.
Computerized scaling dynamically allocates sources as demand grows, mixed with superior safety, resiliency and cost-effective tiered storage, Azure’s storage platform units the tempo for next-generation workloads, delivering the efficiency, scalability and reliability required.
AI WAN: Connecting a number of datacenters for an excellent bigger AI supercomputer
These new AI datacenters are a part of a worldwide community of Azure AI datacenters, interconnected through our Large Space Community (WAN). This isn’t nearly one constructing, it’s a few distributed, resilient and scalable system that operates as a single, highly effective AI machine. Our AI WAN is constructed with progress capabilities in AI-native bandwidth scales to allow large-scale distributed coaching throughout a number of, geographically numerous Azure areas, thus permitting prospects to harness the ability of an enormous AI supercomputer.
This can be a basic shift in how we take into consideration AI supercomputers. As an alternative of being restricted by the partitions of a single facility, we’re constructing a distributed system the place compute, storage and networking sources are seamlessly pooled and orchestrated throughout datacenter areas. This implies larger resiliency, scalability and suppleness for patrons.
Bringing all of it collectively
To satisfy the essential wants of the biggest AI challenges, we wanted to revamp each layer of our cloud infrastructure stack. This isn’t nearly remoted breakthroughs, however composing a number of new approaches throughout silicon, servers, networks and datacenters, resulting in developments the place software program and {hardware} are optimized as one purpose-built system.
Microsoft’s Wisconsin datacenter will play a essential position in the way forward for AI, constructed on actual expertise, actual funding and actual neighborhood impression. As we join this facility with different regional datacenters, and as each layer of our infrastructure is harmonized as an entire system, we’re unleashing a brand new period of cloud-powered intelligence, safe, adaptive and prepared for what’s subsequent.
To study extra about Microsoft’s datacenter improvements, try the digital datacenter tour at datacenters.microsoft.com.
Scott Guthrie is liable for hyperscale cloud computing options and providers together with Azure, Microsoft’s cloud computing platform, generative AI options, knowledge platforms and data and cybersecurity. These platforms and providers assist organizations worldwide clear up pressing challenges and drive long-term transformation.
