Microsoft Outfits Azure Cloud for Big Compute

By Tiffany Trader

November 13, 2012

On Tuesday at SC12 Microsoft debuted a set of “big compute” capabilities for its Windows Azure offering. The company is courting the HPC space with more powerful hardware, new instance configurations, and the updated Microsoft HPC Pack 2012. The advanced management software has many new features and supports the running of compute-intensive workloads in three configurations: on premise, on Windows Azure, or a mixed use, hybrid scenario.

Big Compute on Windows Azure – it’s unclear whether that’s the final name – is available in two configurations. The entry-level HPC instance sports 8 cores and 60 GB RAM, while the higher-end option doubles these specs for a total of 16 cores and 120 GB of RAM. The new servers are outfitted with dual Xeon E5-2670s (2.6 GHz) with DDR3 1600 MHz RAM. The nodes are linked via a 40 Gbps InfiniBand network with RDMA, while a 10GbE backplane is used to hook up to external storage and the Internet.

The new configurations call to mind Amazon’s High-Memory Instances. The High-Memory Double Extra Large Instance (m2.2xlarge) has 4 virtual cores with 34 GB of memory, while the High-Memory Quadruple Extra Large Instance (m2.4xlarge) has 8 virtual cores and 68.4 GB of memory.

An important distinction, however, is that Amazon is using 10GbE interconnect technology – and then only in its Cluster Compute and Cluster GPU instances – while Microsoft is introducing the ability to do RDMA (remote direct memory access) in a virtualized environment. The technology provides low-latency network capability for MPI (message passing interface) applications and allows an Azure cluster to send a 4 byte packet across machines in 2.1 microseconds. Alex Sutton, group program manager for Windows HPC Server, interviewed for this article at SC12, said that Microsoft is the first company to offer virtualized RDMA in a commercial environment.

“For applications written to use the message passing interface (MPI) library, RDMA allows memory on multiple computers to act as one pool of memory,” writes Bill Hilf, general manager, Windows Azure Product Marketing, in a blog entry. “Our RDMA solution provides near bare metal performance (i.e., performance comparable to that of a physical machine) in the cloud, which is especially important for Big Compute applications.”

According to Sutton, the performance penalty for running virtualization is down to about a 1 or 2 percent difference now. This will appeal to organizations that want to access the benefits of cloud (flexibility, scalability, on-demand, etc.), but aren’t willing to sacrifice performance. The use of InfiniBand also enhances throughput, allowing applications to scale more effectively and improving time to results.

As a proof of concept, Microsoft ran the LINPACK benchmark across 504 16-core virtual machines (8,064 cores total). The test cluster, named Faenov, achieved 151.3 teraflops (167.7 peak) with 90.2 percent efficiency, earning it the 165th spot on the most recent TOP500 list. In terms of efficiency, the system placed 27th. Faenov ran Windows Server 2012 in virtual machines hosted on Windows Azure on top of Hyper-V virtualization. Sutton makes the point that 90.2 percent efficiency is better than many on-premise (non-virtualized) clusters.

Bringing system I/O latency under control still leaves the bandwidth barrier that is the consumer Internet, but for the majority of customers, this won’t be an issue. For those that need to make large data transfer into and out of the cloud, Microsoft plans to support “FedEx net,” (physical shipping of drives) at some point.

Pricing on the new configurations has not been announced, so price point comparisons to EC2, Google Compute Engine and other IaaS offerings won’t be possible yet. Initially “Big Compute” will only run Windows, but they are looking into Linux. Of course, the hardware can support Linux, but the engineers still need to hammer out how to run it on virtualized RDMA.

Microsoft is describing early success stories around a segment of customers who run Windows and need low-latency. Initial interest and customer stories are in the areas of risk modeling, disease research and complex engineering tasks. Big data is also on Microsoft’s radar, as the company anticipates many big data workloads benefiting from the new configurations.

Today’s announcement shows us a Microsoft that continues to evolve on the cloud front, both to compete against EC2 and in its support for the HPC community. Azure was originally launched as a PaaS offering in 2010, but in June of this year, Microsoft added infrastructure as a service (IaaS) capabilities and began allowing users to spin up Linux VMs. Customers want choice, but with its purpose-built architecture and significant lead time, Amazon is going to be tough to catch. Microsoft has a dedicated following of Windows users, but most of the action in the HPC community is around Linux.

It will be interesting to see whether low-latency virtualization pans out as a differentiator for Azure. It might take some R&D work, but Amazon could similarly outfit their cloud if they see a call for it. In order for the cloud to be profitable, it has to maintain the right balance of utilization. Too much extra inventory is as bad for business in the long run as too little inventory is in the short run. Cloud companies want just the right about of cushion (or excess inventory). To this point, Microsoft says that it is tracking demand and keeping tight control on the ordering process.

Big Compute on Windows Azure is currently in private preview with select partners. A public beta period is expected to commence in the first half of 2013, followed by general availability in roughly the same time frame.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

Quantum Watchers – Terrific Interview with Caltech’s John Preskill by CERN

July 17, 2024

In case you missed it, there's a fascinating interview with John Preskill, the prominent Caltech physicist and pioneering quantum computing researcher that was recently posted by CERN’s department of experimental physi Read more…

Aurora AI-Driven Atmosphere Model is 5,000x Faster Than Traditional Systems

July 16, 2024

While the onset of human-driven climate change brings with it many horrors, the increase in the frequency and strength of storms poses an enormous threat to communities across the globe. As climate change is warming ocea Read more…

Researchers Say Memory Bandwidth and NVLink Speeds in Hopper Not So Simple

July 15, 2024

Researchers measured the real-world bandwidth of Nvidia's Grace Hopper superchip, with the chip-to-chip interconnect results falling well short of theoretical claims. A paper published on July 10 by researchers in the U. Read more…

Belt-Tightening in Store for Most Federal FY25 Science Budets

July 15, 2024

If it’s summer, it’s federal budgeting time, not to mention an election year as well. There’s an excellent summary of the curent state of FY25 efforts reported in AIP’s policy FYI: Science Policy News. Belt-tight Read more…

Peter Shor Wins IEEE 2025 Shannon Award

July 15, 2024

Peter Shor, the MIT mathematician whose ‘Shor’s algorithm’ sent shivers of fear through the encryption community and helped galvanize ongoing efforts to build quantum computers, has been named the 2025 winner of th Read more…

Weekly Wire Roundup: July 8-July 12, 2024

July 12, 2024

HPC news can get pretty sleepy in June and July, but this week saw a bump in activity midweek as Americans realized they still had work to do after the previous holiday weekend. The world outside the United States also s Read more…

Aurora AI-Driven Atmosphere Model is 5,000x Faster Than Traditional Systems

July 16, 2024

While the onset of human-driven climate change brings with it many horrors, the increase in the frequency and strength of storms poses an enormous threat to com Read more…

Shutterstock 1886124835

Researchers Say Memory Bandwidth and NVLink Speeds in Hopper Not So Simple

July 15, 2024

Researchers measured the real-world bandwidth of Nvidia's Grace Hopper superchip, with the chip-to-chip interconnect results falling well short of theoretical c Read more…

Shutterstock 2203611339

NSF Issues Next Solicitation and More Detail on National Quantum Virtual Laboratory

July 10, 2024

After percolating for roughly a year, NSF has issued the next solicitation for the National Quantum Virtual Lab program — this one focused on design and imple Read more…

NCSA’s SEAS Team Keeps APACE of AlphaFold2

July 9, 2024

High-performance computing (HPC) can often be challenging for researchers to use because it requires expertise in working with large datasets, scaling the softw Read more…

Anders Jensen on Europe’s Plan for AI-optimized Supercomputers, Welcoming the UK, and More

July 8, 2024

The recent ISC24 conference in Hamburg showcased LUMI and other leadership-class supercomputers co-funded by the EuroHPC Joint Undertaking (JU), including three Read more…

Generative AI to Account for 1.5% of World’s Power Consumption by 2029

July 8, 2024

Generative AI will take on a larger chunk of the world's power consumption to keep up with the hefty hardware requirements to run applications. "AI chips repres Read more…

US Senators Propose $32 Billion in Annual AI Spending, but Critics Remain Unconvinced

July 5, 2024

Senate leader, Chuck Schumer, and three colleagues want the US government to spend at least $32 billion annually by 2026 for non-defense related AI systems.  T Read more…

Point and Click HPC: High-Performance Desktops

July 3, 2024

Recently, an interesting paper appeared on Arvix called Use Cases for High-Performance Research Desktops. To be clear, the term desktop in this context does not Read more…

Atos Outlines Plans to Get Acquired, and a Path Forward

May 21, 2024

Atos – via its subsidiary Eviden – is the second major supercomputer maker outside of HPE, while others have largely dropped out. The lack of integrators and Atos' financial turmoil have the HPC market worried. If Atos goes under, HPE will be the only major option for building large-scale systems. Read more…

Everyone Except Nvidia Forms Ultra Accelerator Link (UALink) Consortium

May 30, 2024

Consider the GPU. An island of SIMD greatness that makes light work of matrix math. Originally designed to rapidly paint dots on a computer monitor, it was then Read more…

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock_1687123447

Nvidia Economics: Make $5-$7 for Every $1 Spent on GPUs

June 30, 2024

Nvidia is saying that companies could make $5 to $7 for every $1 invested in GPUs over a four-year period. Customers are investing billions in new Nvidia hardwa Read more…

Nvidia Shipped 3.76 Million Data-center GPUs in 2023, According to Study

June 10, 2024

Nvidia had an explosive 2023 in data-center GPU shipments, which totaled roughly 3.76 million units, according to a study conducted by semiconductor analyst fir Read more…

AMD Clears Up Messy GPU Roadmap, Upgrades Chips Annually

June 3, 2024

In the world of AI, there's a desperate search for an alternative to Nvidia's GPUs, and AMD is stepping up to the plate. AMD detailed its updated GPU roadmap, w Read more…

Some Reasons Why Aurora Didn’t Take First Place in the Top500 List

May 15, 2024

The makers of the Aurora supercomputer, which is housed at the Argonne National Laboratory, gave some reasons why the system didn't make the top spot on the Top Read more…

Intel’s Next-gen Falcon Shores Coming Out in Late 2025 

April 30, 2024

It's a long wait for customers hanging on for Intel's next-generation GPU, Falcon Shores, which will be released in late 2025.  "Then we have a rich, a very Read more…

Leading Solution Providers

Contributors

Google Announces Sixth-generation AI Chip, a TPU Called Trillium

May 17, 2024

On Tuesday May 14th, Google announced its sixth-generation TPU (tensor processing unit) called Trillium.  The chip, essentially a TPU v6, is the company's l Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

IonQ Plots Path to Commercial (Quantum) Advantage

July 2, 2024

IonQ, the trapped ion quantum computing specialist, delivered a progress report last week firming up 2024/25 product goals and reviewing its technology roadmap. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

The NASA Black Hole Plunge

May 7, 2024

We have all thought about it. No one has done it, but now, thanks to HPC, we see what it looks like. Hold on to your feet because NASA has released videos of wh Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire