Mellanox Poised For HDR InfiniBand Quantum Leap

InfiniBand and Ethernet are in a game of tug of war and are pushing the bandwidth and price/performance envelopes constantly. But the one thing they cannot do is get too far out ahead of the PCI-Express bus through which network interface cards hook into processors. The 100 Gb/sec links commonly used in Ethernet and InfiniBand server adapters run up against bandwidth ceilings with two ports running on PCI-Express 3.0 slots, and it is safe to say that 200 Gb/sec speeds will really need PCI-Express 4.0 slots to have two ports share a slot.

This, more than any other factor, is …

Mellanox Poised For HDR InfiniBand Quantum Leap was written by Timothy Prickett Morgan at The Next Platform.

Top 500 Supercomputer Rankings Losing Accuracy Despite High Precision

If the hyperscalers have taught us anything, it is that more data is always better. And because of this, we have to start out by saying that we are grateful to the researchers who created and administer the Top 500 supercomputer benchmark tests for the past 25 years, creating an astonishing 50 consecutive lists ranking the most powerful machines in the world as gauged by the double precision Linpack Fortran parallel matrix math test.

This set of data stands out among a few other groups of benchmarks that have been used by the tens of thousands of organizations – academic …

Top 500 Supercomputer Rankings Losing Accuracy Despite High Precision was written by Timothy Prickett Morgan at The Next Platform.

Nvidia Breaks $2 Billion Datacenter Run Rate

If GPU acceleration had not been conceived of by academics and researchers at companies like Nvidia more than a decade ago, how much richer would Intel be today? How many more datacenters would have had to be expanded or built? Would HPC have stretched to try to reach exascale, and would machine learning have fulfilled the long-sought promise of artificial intelligence, or at least something that looks like it?

These are big questions, and relevant ones, as Nvidia’s datacenter business has just broken through the $2 billion run rate barrier. With something on the order of a 10X speedup across …

Nvidia Breaks $2 Billion Datacenter Run Rate was written by Timothy Prickett Morgan at The Next Platform.

HPC Heavyweight Goes All-In On OpenACC

Across the HPC community, commercial firms, government labs and academic institutions are adapting their code to embrace GPU architectures. They are motivated by the faster performance and lower energy consumption provided by GPUs, and many of them are using OpenACC to annotate their code and make it GPU-friendly. The Next Platform recently interviewed one key organization to learn why it is using the OpenACC programming model to expand its computing capabilities and platform support.

If the earth was the size of a basketball, its atmosphere would be the thickness of shrink wrap. It is fragile enough that in 1960, the …

HPC Heavyweight Goes All-In On OpenACC was written by Timothy Prickett Morgan at The Next Platform.

Qualcomm’s Amberwing Arm Server Chip Finally Takes Flight

It is going to be a busy week for chip maker Qualcomm as it formally jumps from smartphones to servers with its new “Amberwing” Centriq 2400 Arm server processor during the same week that it has received an unsolicited $130 billion takeover offer from sometimes rival chipmaker Broadcom.

The Centriq 2400 is the culmination of over four years of work and investment, which according to the experts in the semiconductor industry we have talked to, easily took on the order of $100 million to $125 million to make happen ­– remember there was a prototype as well as the …

Qualcomm’s Amberwing Arm Server Chip Finally Takes Flight was written by Timothy Prickett Morgan at The Next Platform.

Keeping OpenStack On The Edge, Bleeding And Otherwise

Nobody likes to talk about the scope and scale of platforms than we do at The Next Platform. Almost all of the interesting frameworks for various kinds of distributed computing are open source projects, but the lack of fit and finish is a common complaint across open source software projects.

As Mark Collier, chief operating officer at the OpenStack Foundation, puts it succinctly: “Open source doesn’t have an innovation problem. It has an integration problem.”

Collier’s chief concern, as well as that of his compatriot, Jonathan Bryce, executive director of the OpenStack Foundation and a former Racker ­– meaning …

Keeping OpenStack On The Edge, Bleeding And Otherwise was written by Timothy Prickett Morgan at The Next Platform.

HPE’s Superdome Gets An SGI NUMAlink Makeover

When Hewlett Packard Enterprise bought supercomputer maker SGI back in August 2016 for $275 million, it had already invested years in creating its own “DragonHawk” chipset to build big memory Superdome X systems that were to be the follow-ons to its PA-RISC and Itanium Superdome systems. The Superdome X machines did not support HPE’s own VMS or HP-UX operating systems, but venerable Tandem NonStop fault tolerant distributed database platform was put on the road to Intel’s Xeon processors four years ago.

Now, HPE is making another leap, as we suspected it would, and anointing the SGI UV-300 platform as its …

HPE’s Superdome Gets An SGI NUMAlink Makeover was written by Timothy Prickett Morgan at The Next Platform.

How The Largest Tech Deal In History Might Affect Systems

Private equity firm Silver Lake Partners has an appetite for tech, and securing funding for Dell to take itself private and then go out and buy EMC and VMware is now going to take a backseat in terms of deal size – and in potential ripple effects in the datacenter – now that chip giant Broadcom is making an unsolicited bid, backed by Silver Lake, to take over often-times chip rival Qualcomm.

Should this deal pass shareholder and regulatory, it could finally create a chip giant that can counterbalance Intel in the datacenter – something that Broadcom and Qualcomm both …

How The Largest Tech Deal In History Might Affect Systems was written by Timothy Prickett Morgan at The Next Platform.

For Google Networks, Predictable Latency Trumps Everything

If you want to build infrastructure that scales larger than a single image of a server and an operating system, you have no choice but to network together multiple machines. And so, the network becomes a kind of hyper backplane between compute elements and, in many cases, also a kind of virtual peripheral bus for things like disk and flash storage. From the outside, a warehouse-scale computer, as Google has been calling them for nearly a decade, is meant to look and behave like one machine even if it most certainly is not.

It is hard to quantify how …

For Google Networks, Predictable Latency Trumps Everything was written by Timothy Prickett Morgan at The Next Platform.

Cray Looks Forward To Supercomputing Rebound

The general HPC market might be growing, and the very definition of HPC is expanding thanks to the addition of advanced analytics and machine learning to the HPC toolbox. But it is tough slogging right now in the upper echelons of HPC where supercomputers roam.

There is perhaps no better barometer of the state of supercomputing than Cray, which sells a mix of processing, storage, and interconnect technologies to address the ever-widening scope of modern supercomputing. Because of a general slowdown in supercomputer sales thanks to the fact that organizations are keeping their systems around for longer than they usually …

Cray Looks Forward To Supercomputing Rebound was written by Timothy Prickett Morgan at The Next Platform.

The Tug Of War Between InfiniBand And Ethernet

If you want to get a microcosmic view of the epic battle between Ethernet and InfiniBand (which also includes Omni-Path no matter how much Intel protests) as they relate to high performance computing in its many modern guises, there is perhaps no better place to look at what Mellanox Technologies is selling.

Mellanox, which has been peddling InfiniBand chips, switches, and adapters since the inception of this technology, bought its biggest rival in switch sales, Voltaire, for $218 million back in November 2010. And that was perhaps its smartest move right up to the moment where the company launched …

The Tug Of War Between InfiniBand And Ethernet was written by Timothy Prickett Morgan at The Next Platform.

Intel’s Data Center Group Has Its Head In The Clouds

The cloud gives, and it takes away.

The big hyperscalers, public cloud builders, and telecom, wireless, and cable service providers who are all collectively called “cloud” when it comes to the infrastructure they build, and they are increasingly driving shipments and revenues of all manner of components. But they command, by virtue of their huge volumes, discounts that are much deeper than the typical enterprise customer can get when they buy through an OEM or, if they are large enough, an ODM.

The fact that Intel’s Data Center Group is managing to profit pretty handsomely and reasonably predictably despite this …

Intel’s Data Center Group Has Its Head In The Clouds was written by Timothy Prickett Morgan at The Next Platform.

Can Vector Supercomputing Be Revived?

Seymour Cray loved vector supercomputers, and made the second part of that term a household word because of it. NEC, the last of the pure vector supercomputer makers, is so excited about its new “Aurora” SX-10+ vector processor and the “Tsubasa” supercomputer that will use it that it forgot to announce the processor to the world when it previewed the system this week.

Here at The Next Platform, we easily forgive such putting of carts before horses – so long as someone eventually explains the horse to us before the cart starts shipping for real. NEC is expected to …

Can Vector Supercomputing Be Revived? was written by Timothy Prickett Morgan at The Next Platform.

AWS First Up With Volta GPUs In The Cloud

It must be tough for the hyperscalers that are expanding into public cloud and the public cloud builders that also use their datacenters to run their own businesses to decide whether to hoard all of the new technologies that they can get their hands on for their own benefit, or to make money selling that capacity to others.

For any new, and usually constrained, kind of capacity, such as shiny new “Skylake” Xeon SP processors from Intel or “Volta” Tesla GPU accelerators from Nvidia, it has to be a hard call for Google, Amazon, Microsoft, Baidu, Tencent, and Alibaba to …

AWS First Up With Volta GPUs In The Cloud was written by Timothy Prickett Morgan at The Next Platform.

The Dream Of Software Only Shared Memory Clusters

It is hard to quantify the amount of effort in systems and application software development that has been done precisely because there is no easy, efficient, and cheap way to make a bunch of cheap commodity servers look like one wonking system with a big ole flat memory space that is as easy to program as a PC but which brings to bear all that compute, memory, and I/O of a cluster as a single system image.

We have SMP and NUMA glue chips to do such shared memory clustering in hardware, scaling from two to four and sometimes eight, …

The Dream Of Software Only Shared Memory Clusters was written by Timothy Prickett Morgan at The Next Platform.

HPE Demystifies Deep Learning For Faster Intelligence

Today’s enterprises need deep learning, but most don’t know how to get started. As rising data volumes and evolving industry trends push the limits of traditional IT, the latest innovations are helping them operate faster and smarter—and high performance computing is just the beginning.

Enterprises are deploying robust server platforms to power HPC applications, leveraging optimal performance, reliability, and flexibility to handle increasingly dense workloads. And with these industry-leading tools, modeling and simulation capabilities are rapidly evolving. Artificial intelligence is transforming how we operate and relate to technology. AI allows machines to think and learn like the human brain, while …

HPE Demystifies Deep Learning For Faster Intelligence was written by Timothy Prickett Morgan at The Next Platform.

BSC Builds 21st Century HPC In A 19th Century Cathedral

This summer, the Partnership for Advanced Computing in Europe (PRACE) added to its roster another of the world’s most powerful high performance computing systems. The Barcelona Computing Center’s new MareNostrum 4, delivered by IBM with the help of partners Lenovo and Fujitsu, and fueled by HPC technologies from Intel, will facilitate extensive engineering and scientific research in fields like astrophysics, weather forecasting, and genome research. Nestled within a unique building – the Torre Girona chapel, which fell out of use – the fourth generation MareNostrum system relies on a general purpose cluster working with three specialized clusters to achieve its …

BSC Builds 21st Century HPC In A 19th Century Cathedral was written by Timothy Prickett Morgan at The Next Platform.

The IBM Transformation Can Gather Steam Now

For the past five and a half years, which is not quite an eternity in the IT business but is something akin to a half of a generation or so, IBM’s revenues have been declining, quarter in and quarter out. As has happened many, many times in its more than century of existence, Big Blue, which used to be a peddler of meat slicers, time machines, scales, and punch card tabulators early in its history, has had to constantly evolve and reimagine itself.

The transformation that IBM had to undergo in the late 1980s and early 1990s was a near …

The IBM Transformation Can Gather Steam Now was written by Timothy Prickett Morgan at The Next Platform.

Getting With The Program On Software Defined Networks

If the profit margins are under pressure among the switch and router makers of the world, their chief financial officers can probably place a lot of the blame on Nick McKeown and his several partners throughout the years. And if McKeown is right about what is happening as the network software is increasingly disaggregated from the hardware – what is called software defined networking – they will either have to adapt or be relegated to the dustbins of history.

McKeown cut his teeth after university in the late 1980s at Hewlett Packard Labs in Bristol, England, one of the hotbeds …

Getting With The Program On Software Defined Networks was written by Timothy Prickett Morgan at The Next Platform.

IBM Preps Power9 For AI And HPC Launch, Forges Big NUMA Iron

After a long, long wait and years of anticipation, it looks like IBM is finally getting ready to ship commercial versions of its Power9 chips, and as expected, its first salvo of processors aimed at the datacenter will be aimed at HPC, data analytics, and machine learning workloads.

We are also catching wind about IBM’s Power9-based scale-up NUMA machines, which will debut sometime next year and take on big iron systems based on Intel Xeon SP, Oracle Sparc M8, and Fujitsu Sparc64-XII processors as well as give some competition to IBM’s own System z14 mainframes.

The US Department …

IBM Preps Power9 For AI And HPC Launch, Forges Big NUMA Iron was written by Timothy Prickett Morgan at The Next Platform.