Server processor architectures are trying to break the ties between memory and compute to allow the capacities of each to scale independently of each other, but switching and routing giant Cisco Systems has already done this for a high-end switch chip that looks remarkably like a CPU tuned for network processing.
At the recent Hot Chips conference in Silicon Valley, Jamie Markevitch, a principal engineer at Cisco showed off the guts of an unnamed but currently shipping network processor, something that happens very rarely in the switching and routing racket. With the exception of the upstarts like Barefoot Networks with …
A Rare Peek Inside A 400G Cisco Network Chip was written by Timothy Prickett Morgan at The Next Platform.
We have heard much about the concept of dark silicon but there is a separate, related companion to this idea.
Dark bandwidth is a term that is being bandied about to describe the major inefficiencies of data movement. The idea of this is not unknown or new, but some of the ways the problem is being tackled present new practical directions as the emphasis on system balance over pure performance persists.
As ARM Research architect, Jonathan Beard describes it, the way systems work now is a lot like ordering a tiny watch battery online and having it delivered in a …
Shedding Light on on Dark Bandwidth was written by Nicole Hemsoth at The Next Platform.
Makers of tightly coupled, shared memory machines can make all of the arguments they want about how it is much more efficient and easier to program these NUMA machines than it is to do distributed computing across a cluster of more loosely coupled boxes, but for the most part, the IT market doesn’t care.
Distributed computing, in its more modern implementation of frameworks running on virtual machines or containers – or both – is by far the norm, both in the datacenter and on the public clouds. You don’t have to look any further than the latest server sales statistics …
Custom Server Makers Set The Datacenter Pace was written by Timothy Prickett Morgan at The Next Platform.
At well over $150,000 per appliance, the Volta GPU based DGX appliances from Nvidia, which take aim at deep learning with framework integration and 8 Volta-accelerated nodes linked with NVlink, is set to appeal to the most bleeding edge of machine learning shops.
Nvidia has built its own clusters by stringing several of these together, just as researchers at Tokyo Tech have done with the Pascal generation systems. But one of the first commercial customers for the Volta based boxes is the Center for Clinical Data Science, which is part of the first wave of hospitals set to …
Hospital Captures First Commercial Volta GPU Based DGX-1 Systems was written by Nicole Hemsoth at The Next Platform.
POSIX I/O is almost universally agreed to be one of the most significant limitations standing in the way of I/O performance exascale system designs push 100,000 client nodes.
The desire to kill off POSIX I/O is a commonly beaten drum among high-performance computing experts, and a variety of new approaches—ranging from I/O forwarding layers, user-space I/O stacks, and completely new I/O interfaces—are bandied about as remedies to the impending exascale I/O crisis.
However, it is much less common to hear exactly why POSIX I/O is so detrimental to scalability and performance, and what needs to change to have a suitably …
What’s So Bad About POSIX I/O? was written by Nicole Hemsoth at The Next Platform.
The world of Ethernet switching and routing used to be more predictable than just about any other part of the datacenter, but for the past decade the old adage – ten times the bandwidth for three times the cost – has not held. While 100 Gb/sec Ethernet was launched in 2010 and saw a fair amount of uptake amongst telecom suppliers for their backbones, the hyperscalers decided, quite correctly, that 100 Gb/sec Ethernet was too expensive and opted for 40 Gb/sec instead.
Now, we are sitting on the cusp of the real 100 Gb/sec Ethernet rollout among hyperscalers and enterprise …
Signposts On The Roadmap Out To 10 Tb/sec Ethernet was written by Timothy Prickett Morgan at The Next Platform.
The rivalry between Mesos, Kubernetes, and OpenStack just keeps getting more interesting, and instead of a winner take all situation, it has become more of a take what you need approach. That said, it is looking like Kubernetes is emerging as the de facto standard for container control, even though Google not the first out of the gate in open sourcing Kubernetes and Docker Swam and the full Docker Enterprise are seeing plenty of momentum in the enterprise.
Choice is a good thing for the IT industry, and the good news is that because of architectural choices made by …
Mesos Borgs Google’s Kubernetes Right Back was written by Timothy Prickett Morgan at The Next Platform.
The era of Hewlett Packard Enterprise’s envious – and expensive – desire to become IT software and services behemoth like the IBM of the 1990s and 2000s is coming to a close.
The company has finalized its spinout-merger of substantially all of its software assets to Micro Focus. HPE has already spun out the lion’s share of its outsourcing and consulting businesses to Computer Sciences and even earlier had split from its troublesome PC and very profitable printer businesses. These were spun out together to give the combined HP Inc a chance to live on Wall Street and because PCs …
The Prospects For A Leaner And Meaner HPE was written by Timothy Prickett Morgan at The Next Platform.
It is difficult not to be impatient for the technologies of the future, which is one reason that this publication is called The Next Platform. But those who are waiting for the Gen-Z consortium to deliver a memory fabric that will break the hegemony of the CPU in controlling access to memory and to deepen the memory hierarchy while at the same time flattening memory addressability are going to have to wait a little longer.
About a year longer, in fact, which is a bit further away than the founders of the Gen-Z consortium were hoping when they launched …
Future Interconnects: Gen-Z Stitches A Memory Fabric was written by Timothy Prickett Morgan at The Next Platform.
There is no question that Intel has reached its peak in the datacenter when it comes to compute. For years now, it has had very little direct competition and only some indirect competition for the few remaining RISC upstarts and the threat of the newbies with their ARM architectures.
The question now, as we ponder the “Skylake” Xeon SP processors and their “Purley” platform that launched in July, is this: Is Intel at a local maximum, with another peak off in the distance, perhaps after a decline or perhaps after steady growth or a flat spot, or is this the …
The Huge Premium Intel Is Charging For Skylake Xeons was written by Timothy Prickett Morgan at The Next Platform.
For a large institution playing at the leadership-class supercomputing level, NASA tends to do things a little differently than its national lab and academic peers.
One of the most striking differences between how the space agency views its supercomputing future can be found at the facilities level. Instead of building massive brick and mortar datacenters within a new or existing complex, NASA has taken the modular route, beginning with its Electra supercomputer and in the near future, with a 30 Megawatt-capable new modular installation that can house about a million compute cores.
“What we found is that the modular approach …
NASA Supercomputing Strategy Takes the Road Less Traveled was written by Nicole Hemsoth at The Next Platform.
Sometimes a database is like a collection of wax tablets that you can stack and sort through to update, and these days, sometimes it is more like a river that has a shape defined by its geography but it is constantly changing and flowing and that flow, more than anything else, defines the information that drives the business. There is no time to persist it, organize it, and then query it.
In this case, embedding a database right in that stream makes good sense, and that is precisely what Confluent, the company that has commercialized Apache Kafka, which is a …
Kafka Wakes Up And Is Metamorphosed Into A Database was written by Timothy Prickett Morgan at The Next Platform.
Building a platform is hard enough, and there are very few companies that can build something that scales, supports a diversity of applications, and, in the case of either cloud providers or software or whole system sellers, can be suitable for tens of thousands, much less hundreds of thousands or millions, of customers.
But if building a platform is hard, keeping it relevant is even harder, and those companies who demonstrate the ability to adapt quickly and to move to new ground while holding old ground are the ones that get to make money and wield influence in the datacenter. …
VMware’s Platform Revolves Around ESXi, Except Where It Can’t was written by Timothy Prickett Morgan at The Next Platform.
We continue with our second part of the series on the Tsubame supercomputer (first section here) with the next segment of our interview with Professor Satoshi Matsuoka, of the Tokyo Institute of Technology (Tokyo Tech).
Matsuoka researches and designs large scale supercomputers and similar infrastructures. More recently, he has worked on the convergence of Big Data, machine/deep learning, and AI with traditional HPC, as well as investigating the post-Moore technologies towards 2025. He has designed supercomputers for years and has collaborated on projects involving basic elements for the current and more importantly future exascale systems.
TNP: Will you be running …
Heterogeneous Supercomputing on Japan’s Most Powerful System was written by Nicole Hemsoth at The Next Platform.
According to a recent Jefferies report, the fourth wave of computing has started and it is being driven by the adoption of IoT with parallel processing as the solution. Tectonic shifts in computing have been caused by major forces dating back to the 1960s.
With each shift, new solution providers have emerged as prominent suppliers. The latest power often cited with the fourth wave is Nvidia and its parallel processing platform for HPC and artificial intelligence (AI), namely GPUs and the CUDA programming platform. The growth of the data center segment of Nvidia’s business – from $339 million in …
The Rise Of The Fourth Wave Of Computing was written by Timothy Prickett Morgan at The Next Platform.
There are a lot of different ways to skin the deep learning cat. But for hyperscalers and cloud providers who want to use a single platform internally as well as providing deep learning services to customers externally, they really want to have as few different architectures as possible in their datacenters to maximize efficiencies and to lower both capital and operational costs. This is particularly true when the hyperscaler is also a cloud provider.
If Moore’s Law had not run out of gas – or at least shifted to lower octane fuel – then the choice would have been easy. …
Drilling Into Microsoft’s BrainWave Soft Deep Learning Chip was written by Timothy Prickett Morgan at The Next Platform.
Many have tried to wrench the door of the datacenter open with ARM processors, but Qualcomm, which knows a thing or two about creating and selling chips for smartphones and other client devices, has perhaps the best chance of actually selling ARM chips in volume inside of servers.
The combination of a rich and eager target market with a good product design tailored for that market and enough financial strength and stability to ensure many generations of development are what are necessary to break into the datacenter, and the “Falkor” cores that were unveiled this week at Hot Chips were …
ARM Servers: Qualcomm Is Now A Contender was written by Timothy Prickett Morgan at The Next Platform.
Propping up a successful silicon startup is no simple feat, but venture-backed Wave Computing has managed to hold its own in the small but critical AI training chip market–so far.
Seven years after its founding and the company’s early access program for beta machines based on its novel DPU manycore architecture is now open, which is prompting Wave to be more forthcoming about the system and chip architecture for deep learning-focused dataflow architecture.
Dr. Chris Nicol, Wave Computing CTO and lead architect of the Dataflow Processing Unit (DPU) admitted to the crowd at Hot Chips this week that maintaining funding …
First In-Depth View of Wave Computing’s DPU Architecture, Systems was written by Nicole Hemsoth at The Next Platform.
Professor Satoshi Matsuoka, of the Tokyo Institute of Technology (Tokyo Tech) researches and designs large scale supercomputers and similar infrastructures. More recently, he has worked on the convergence of Big Data, machine/deep learning, and AI with traditional HPC, as well as investigating the Post-Moore Technologies towards 2025.
He has designed supercomputers for years and has collaborated on projects involving basic elements for the current and more importantly future Exascale systems. I talked with him recently about his work with the Tsubame supercomputers at Tokyo Tech. This is the first in a two-part article. For background on the Tsubame 3 system …
Inside View: Tokyo Tech’s Massive Tsubame 3 Supercomputer was written by Nicole Hemsoth at The Next Platform.
In the U.S. it is easy to focus on our native hyperscale companies (Google, Amazon, Facebook, etc.) and how they design and deploy infrastructure at scale.
But as our regular readers understand well, the equivalent to Google in China, Baidu, has been at the bleeding edge with chips, systems, and software to feed its own cloud-delivered and research operations.
We’ve written much over the last few years about the company’s emphasis on streamlining deep learning processing, most notably with GPUs, but Baidu has a new processor up its sleeve called the XPU. For now, the device has just been demonstrated …
An Early Look at Baidu’s Custom AI and Analytics Processor was written by Nicole Hemsoth at The Next Platform.