Tag Archives: InfiniBand

Paving The Road to Exascale – Part 2 of many

In the introduction for the “Paving the road to Exascale” series of posts (part 1), one of the items I mentioned was the “many many cores, CPU or GPUs”.  The basic performance of a given system is being measured by flops. Each CPU/GPU is capable for X amount of flops (which can be calculated as number of parallel operations * frequency * cores for example), and the sum of all of them in a given system gives you the maximum compute capability of the system. How much you can really utilize for your application depends on the system design, memory bandwidth, interconnect etc. On the Top500 list, you can see per each of the systems listed, what the maximum amount of flops is, and what is the effective or measured performance using the Linpack benchmark.

In order to achieve the increasing performance targets (we are talking on paving the road to Exascale….) we need to have as many cores as possible. As we all witness, GPUs have become the most cost-effective compute element, and the natural choice for bringing the desired compute capability in the next generation of supercomputers. A simple comparison shows that with a proprietary design, such as a Cray machine, one needs around 20K nodes to achieve Petascale computing, and using GPUs (assuming one per server). 5K nodes will be enough to achieve a similar performance capability – best cost effective solution.

So, now that we starting to plug in more and more GPUs into the new supercomputers, there are two things that we need to take care of – one, is to start working on the application side, and port applications to use parallel GPU computation (this is a subject for a whole new blog) and second, to make sure the communications between the GPU is as effective as possible. For the later, we have saw the recent announcements from NVIDIA and Mellanox on creating a new interface, called GPUDirect, that enables a better and more efficient communication interface between the GPUs and the InfiniBand interconnect. The new interface eliminates the CPU involvement from the GPU communications data path, using the host memory as the medium between the GPU and the InfiniBand adapter. One needs to be aware, that the GPUDirect solution requires network offloading capability to completely eliminate the CPU from being involved in the data path, as if the network requires CPU cycles to send and receive traffic, the CPU will still be involved in the data path! Once you eliminate the CPU from the GPU data path, you can reduce the GPU communications by 30%.

We will be seeing more and more optimizations for GPU communications on high speed networks. The end goal is of course to provide local system latencies for remote GPUs, and with that ensure the maximum utilization of the GPU’s flops capability.

Till next time,

Gilad Shainer

The biggest winner of the new June 2010 Top500 Supercomputers list? InfiniBand!

Published twice a year, the Top500 supercomputers list ranks the world fastest supercomputers and provides a great indication for HPC market trends, usage models and a tool for future predictions. The 35th release of the Top500 list was just published and according to the new results InfiniBand has become the de-facto interconnect technology for high performance computing.

What wasn’t said on InfiniBand from the competitor world? Too many time I have heard that InfiniBand is dead and that Ethernet is the killer. I am just sitting in my chair and laughing. InfiniBand is the only interconnect that is growing on the Top500 list, more than 30% growth year over year (YoY) and it is growing by continuing to uproot Ethernet and the proprietary solutions. Ethernet is 14% down YoY and it has become very difficult to spot a proprietary clustered interconnect…  Even more, in the hard core of HPC, the Top100, 64% of the systems are InfiniBand and are using solutions from Mellanox. InfiniBand is definitely proven to provide the needed scalability, efficiency and performance, and to really deliver the highest CPU or GPU availability to the user or to the applications. Connecting 208 systems from the list is only steps away from connecting the majority of the systems.

What makes InfiniBand so strong? The fact that it solves issues and does not migrate them to other parts of the systems. In a balanced HPC system, each components needs to do its work, and not rely on other components to do overhead tasks. Mellanox is doing a great job in providing solutions that offload all the communications and can provide the needed accelerations for the CPU or GPU, and maximize the CPU/GPU cycles for the applications. The collaborations with NVIDIA on the NVIDA GPUDirect, Mellanox CORE-Direct and so forth are just few examples.

The GPUDIrect is a great example on how Mellanox can offload the CPU from being involved in the GPU-to-GPU communications. No other InfiniBand vendor can do it without using Mellanox technology. GPUDirect requires network offloading or it does not work. Simple. When you want to offload the CPU from being involved in the GPU to GPU communications, and your interconnect needs the CPU to do the transports (since it is an onloading solution), the CPU is involved in every GPU transaction. Only offloading interconnects, such as Mellanox InfiniBand can really deliver the benefits of the GPUDirect.

If you want more information on the GPUDirect and other solutions, feel free to drop a note to hpc@mellanox.com.


Visit Mellanox at ISC’10

It’s almost time for ISC’10 in Hamburg, Germany (May 31-June 3), please stop by and visit Mellanox Technologies booth (#331) to learn more about how our products deliver market-leading bandwidth, high-performance, scalability, power conservation and cost-effectiveness while converging multiple legacy network technologies into one future-proof solution.  

Mellanox’s end-to-end 40Gb/s InfiniBand connectivity products deliver the industry’s leading CPU efficiency rating on the TOP500. Come see our application acceleration and offload technologies that decrease run time and increase cluster productivity.

Hear from our HPC Industry Exports

Exhibitor Forum Session – Tuesday, June 1, 9:40AM – 10:10AM

Speaking: Gilad Shainer, Sr. Director of HPC Marketing / Michael Kagan, CTO

HOT SEAT SESSION – Tuesday, June 1, 3:15PM – 3:30PM

Speaking: Michael Kagan, CTO

JuRoPa breakfast Session – Wednesday, June 2, 7:30AM – 8:45AM

Speaking: Gilad Shainer, Sr. Director of HPC Marketing / Michael Kagan, CTO

“Low Latency, High Throughput, RDMA & the Cloud In-Between” – Wednesday, June 2, 10:00AM – 10:30AM

Speaking: Gilad Shainer, Sr. Director of HPC Marketing

“Collectives Offloads for Large Scale Systems” – Thursday, June 3, 11:40AM – 12:20PM

Speaking: Gilad Shainer, Mellanox Technologies; Prof. Dr. Richard Graham, Oak Ridge National Laboratory

“RoCE – New Concept of RDMA over Ethernet” – Thursday, June 3, 12:20PM – 1:00PM

Speaking: Gilad Shainer, Sr. Director of HPC Marketing and Bill Lee, Sr. Product Marketing Manager

Paving The Road to Exascale – Part 1 of many

1996 was the year when the world saw the first Teraflops system. 12 years after, the first Petaflop system was built. It took the HPC world 12 years to increase the performance by a factor of 1000. Exascale computing, another performance jump by a factor of 1000 will not take another 12 years. Expectations indicate that we will see the first Exascale system in the year 2018, only 10 years after the introduction of the Petaflop system. How do we get to the Exascale system is a good question, but we definitely put some guidelines on how to do it right. Since there is much to write on this subject, this will probably take multiple blog posts, and we have time till 2018…  :)

Here are the items that I have in mind as overall guidelines:

-  Dense computing – we can’t populate Earth with servers as we need some space for living… so dense solutions will need to be built – packing as many cores as possible in a single rack. This is a task for the Dell folks…  :)

-  Power efficiency – energy is limited, and today data centers already consume too much power. Apart from alternative energy solutions, the Exascale systems will need to be energy efficient, and this covers all of the systems components – CPUs, memory, networking. Every Watt is important.

-  Many-many cores – CPU / GPU, as much as possible and be sure, software will use them all

-  Offloading networks – every Watt is important, every flop needs to be efficient. CPU/GPU availability will be critical in order to achieve the performance goals. No one can afford wasting cores on non-compute activities.

-  Efficiency – balanced systems, no jitters, no noise, same order of magnitude of latency everywhere – between CPUs, between GPUs, between end-points

-  Ecosystem/partnership is a must – no one can do it by himself.

In future posts I will expand on the different guidelines, and definitely welcome your feedback.

Gilad Shainer
Senior Director, HPC and Technical Computing

Partners Healthcare Cuts Latency of Cloud-based Storage Solution Using Mellanox InfiniBand Technology

Interesting article just came out from Dave Raffo at SearchStorage.com. I have a quick summary below but you should certainly read the full article here: “Health care system rolls its own data storage ‘cloud’ for researchers.”

Partners HealthCare, a non-profit organization founded in 1994 by Brigham and Women’s Hospital and Massachusetts General Hospital, is an integrated health care system that offers patients a continuum of coordinated high-quality care.

Over the past few years, ever-increasing advances in the resolution and accuracy of medical devices and instrumentation technologies have led to an explosion of data in biomedical research. Partners recognized early on that a Cloud-based research compute and storage infrastructure could be a compelling alternative for their researchers. Not only would it enable them to distribute costs and provide storage services on demand, but it would save on IT management time that was spent fixing all the independent research computers distributed across the Partners network.

Initially, Partners Healthcare chose Ethernet as the network transport technology. As demand grew the solution began hitting significant performance bottlenecks, particularly during read/write of 100’s of thousands of small files. The issue was found to lie with the interconnect—Ethernet created problems due to its high natural latency. In order to provide a scalable low latency solution, Partners Healthcare turned to InfiniBand. With InfiniBand on the storage back end, Partners experienced roughly two orders of magnitude faster read times. “One user had over 1,000 files, but only took up 100 gigs or so,”said Brent Richter corporate manager for enterprise research infrastructure and services, Partners HealthCare System.”Doing that with Ethernet would take about 40 minutes just to list that directory. With InfiniBand, we reduced that to about a minute.”

Also, Partners chose InfiniBand over 10-Gigabit Ethernet because InfiniBand is a lower latency protocol. “InfiniBand was price competitive and has lower latency than 10-Gig Ethernet,” he said.

Richter said the final price tag came to about $1 per gigabyte.

By integrating Mellanox InfiniBand into the storage solution, Partners Healthcare was able to reduce latency close to zero and increase its performance, providing their customers with faster response and higher capacity.

Till next time,

Brian Sparks

Sr. Director, Marketing Communication

TOP500 33rd List Highlights

Started in 1993, the TOP500 lists the fastest computers used today, ranked according to Linpack benchmark results. Published twice a year, the TOP500 list provides an important tool for tracking usage trends in high-performance computing. The 33rd TOP500 List was released in Hamburg, Germany, during the ISC’09 conference.

This year’s list revealed that Mellanox InfiniBand demonstrated up to 94 percent system utilization, only 6 percent under the theoretical limit, providing users with the best return on investment for their high-performance computing server and storage infrastructure. This year’s TOP500 list reveals that InfiniBand is the only growing industry-standard interconnect solution, increasing 25 percent to 152 systems, representing more than 30 percent of the TOP500. Mellanox ConnectX® InfiniBand adapters and switch systems based on its InfiniScale® III and IV switch silicon provide the scalable, low-latency, and power-efficient interconnect for the world’s fastest supercomputer and the majority of the top 100 systems. Mellanox end-to-end 40Gb/s InfiniBand solutions deliver the leading performance and highest Top10 system efficiency in the 10th ranked Jülich cluster.

Highlights of InfiniBand usage on the June 2009 TOP500 list include:

- Mellanox InfiniBand interconnect products connect the world’s fastest supercomputer, 4 of the top 10 most prestigious positions, and 9 of the top 20 systems
- Mellanox InfiniBand provides the highest system utilization, up to 94 percent, which is 50 percent higher than the best GigE-based system
- Mellanox 40Gb/s end-to-end solutions provide the highest system utilization on the top 10, 15% higher than the average top 20 efficiency
- All InfiniBand-based clusters (152 total supercomputers) use Mellanox solutions
- InfiniBand is the most used interconnect among the top 100 supercomputers with 59 systems, nearly 12 times the number of Gigabit Ethernet-based clusters and the number of proprietary high speed cluster interconnects
- The total number of InfiniBand-connected CPU cores on the list has grown from 606,000 in June 2008 to 1,040,000 in 2009 (72 percent yearly growth)
- InfiniBand is the only growing industry-standard clustered interconnect in the TOP500 with a 25 percent growth rate compared to June 2008
- Mellanox InfiniBand interconnect products present in the TOP500 are used by a diverse list of applications, from large-scale, high-performance computing to commercial technical computing and enterprise data centers
- The entry level for the TOP500 list is 17TFlops, 91% percent higher than the 8.9TFlops necessary to be on the June 2008 list

Full analysis of the TOP500 can be found HERE.

Inauguration of 1st European Petaflop Computer in Jülich, Germany

On Tuesday, May 26, the Research Center Jülich reached a significant milestone of German and European supercomputing with the inauguration of two new supercomputers: the supercomputer JUROPA and the fusion machine HPC FF. The symbolic start of the systems were triggered by the German Federal Minister for Education and Research, Prof. Dr. Annette Schavan, the Prime Minister of North Rhine-Westphalia, Dr. Jürgen Rüttgers, and Prof. Dr. Achim Bachem, Chairman of the Board of Directors at Research Center Jülich as well as high-ranking international guests from academia, industry and politics.

JUROPA (which stands for Juelich Research on Petaflop Architectures) will be used Pan-European-wide by more than 200 research groups to run their data-intensive applications. JUROPA is based on a cluster configuration of Sun Blade servers, Intel Nehalem processors, Mellanox 40Gb/s InfiniBand and Cluster Operation Software ParaStation from ParTec Cluster Competence Center GmbH. The system was jointly developed by experts of the Jülich Supercomputing Center and implemented with partner companies Bull, Sun, Intel, Mellanox and ParTec. It consists of 2,208 compute nodes with a total computing power of 207 Teraflops and was sponsored by the Helmholtz Community. Prof. Dr. Dr. Thomas Lippert, Head of Jülich Supercomputing Center, explains the HPC Installation in Jülich in the video below.

HPC-FF (High Performance Computing – for Fusion), drawn up by the team headed by Dr. Thomas Lippert, director of the Jülich Supercomputing Centre, was optimized and implemented together with the partner companies Bull, SUN, Intel, Mellanox and ParTec. This new best-of-breed system, one of Europe’s most powerful, will support advanced research in many areas such as health, information, environment, and energy. It consists of 1,080 computing nodes each equipped with two Nehalem EP Quad Core processors from Intel. Their total computing power of 101 teraflop/s corresponds, at the present moment, to 30th place in the list of the world’s fastest supercomputers. The combined cluster will achieve 300 teraflops/s computing power and will be included in the rating of the Top500 list, published this month at ISC’09 in Hamburg, Germany.

40Gb/s InfiniBand from Mellanox is used as the system interconnect. The administrative infrastructure is based on NovaScale R422-E2 servers from French supercomputer manufacturer Bull, who supplied the compute hardware and the SUN ZFS/Lustre Filesystem. The cluster operating system “ParaStation V5″ is supplied by Munich software company ParTec. HPC-FF is being funded by the European Commission (EURATOM), the member institutes of EFDA, and Forschungszentrum Jülich.

Complete System facts: 3288 compute nodes ; 79 TB main memory; 26304 cores; 308 Teraflops peak performance

Missed Mellanox at Interop?

Just in case you missed us at Interop 2009, below are just a few of the presentations that took place in our booth.

Mellanox 10 Gigabit Ethernet and 40Gb/s InfiniBand adapters, switches and gateways are key to making your data center F.U.E.L. Efficient


Mellanox Product Manager, Satish Kikkeri, provides additional details on Low-Latency Ethernet


Mellanox Product Manager, TA Ramanujam, provides insight on how data centers can achieve true unified I/O today


Fusion-io’s CTO, David Flynn, presents “Moving Storage to Microsecond Time-Scales”


We look forward to seeing you at our next event or tradeshow.

Brian Sparks

Mellanox Key to Fusion-io’s Demo at Interop

I’m still pondering my take on Interop this year. It’s been a while since I’ve seen so many abandoned spaces on the show floor. Mind you most were 10×10 or 10×20 spots, but you could tell there were others who really went light on their presence. I saw one booth which had a 40×40 booth and just filled it with banner stands. Yikes! So nothing was really grabbing at me until I went to Fusion-io’s booth and saw the wall of monitors with a 1,000 videos playing on it at once. Fusion-io Booth


Even more amazing, the videos were all being driven by a single PCIe card which had 1.2TB of SSD RAM on it. This one “ioSAN” card from Fusion-io completely saturated 16 servers (126 cpu cores)…and they were able to achieve this through the bandwidth performance and ultra low-latency of 20Gb/s InfiniBand via Mellanox’s ConnectX adapters. In fact, they told me the 20Gb/s InfiniBand connection would allow them to saturate even more servers, yet they only brought 16.


The video below, featuring Fusion-io’s CTO David Flynn, tells the complete story:

The ioSAN can be used as networked, server-attached storage or integrated into networked storage infrastructure, making fundamental changes to the enterprise storage area. This is a great example of how Mellanox InfiniBand is the enabling technology for next generation storage.

Talk with you again soon,

Brian Sparks