Archive

Category Archives for "The Next Platform"

Google Takes Unconventional Route with Homegrown Machine Learning Chips

At the tail end of Google’s keynote speech at its developer conference Wednesday, Sundar Pichai, Google’s CEO mentioned that Google had built its own chip for machine learning jobs that it calls a Tensor Processing Unit, or TPU.

The boast was that the TPU offered “an order of magnitude” improvement in the performance per watt for machine learning. Any company building a custom chip for a dedicated workload is worth noting, because building a new processor is a multimillion-dollar effort when you consider hiring a design team, the cost of getting a chip to production and building the hardware and

Google Takes Unconventional Route with Homegrown Machine Learning Chips was written by Nicole Hemsoth at The Next Platform.

IBM Extends GPU Cloud Capabilities, Targets Machine Learning

As we have noted over the last year in particular, GPUs are set for another tsunami of use cases for server workloads in high performance computing and most recently, machine learning.

As GPU maker Nvidia’s CEO stressed at this year’s GPU Technology Conference, deep learning is a target market, fed in part by a new range of their GPUs for training and executing deep neural networks, including the Tesla M40, M4, the existing supercomputing-focused K80, and now, the P100 (Nvidia’s latest Pascal processor, which is at the heart of a new appliance specifically designed for deep learning workloads).

While

IBM Extends GPU Cloud Capabilities, Targets Machine Learning was written by Nicole Hemsoth at The Next Platform.

Climate Research Pulls Deep Learning Onto Traditional Supercomputers

Over the last year, stories pointing to a bright future for deep neural networks and deep learning in general have proliferated. However, most of what we have seen has been centered on the use of deep learning to power consumer services. Speech and image recognition, video analysis, and other features have spun from deep learning developments, but from the mainstream view, it would seem that scientific computing use cases are still limited.

Deep neural networks present an entirely different way of thinking about a problem set and the data that feeds it. While there are established approaches for images and

Climate Research Pulls Deep Learning Onto Traditional Supercomputers was written by Nicole Hemsoth at The Next Platform.

In-Memory Breathes New Life Into NUMA

Hyperscalers and the academics that often do work with them have invented a slew of distributed computing methods and frameworks to get around the problem of scaling up shared memory systems based on symmetric multiprocessing (SMP) or non-uniform memory access (NUMA) techniques that have been in the systems market for decades. SMP and NUMA systems are expensive and they do not scale to hundreds or thousands of nodes, much less the tens of thousands of nodes that hyperscalers require to support their data processing needs.

It sure would be convenient if they did. But for those who are not hyperscalers,

In-Memory Breathes New Life Into NUMA was written by Timothy Prickett Morgan at The Next Platform.

IBM Throws Weight Behind Phase Change Memory

There is no question that the memory hierarchy in systems is being busted wide open and that new persistent memory technology that can be byte addressable like DRAM or block addressable like storage are going to radically change the architecture of machines and the software that runs on them. Picking what memory might go mainstream is another story.

It has been decades since IBM made its own DRAM, but the company still has a keen interest in doing research and development on core processing and storage technologies and in integrating new devices with its Power-based systems.

To that end, IBM

IBM Throws Weight Behind Phase Change Memory was written by Timothy Prickett Morgan at The Next Platform.

Scaling All Flash Arrays Up And Out

The ubiquity of the Xeon server has been a boon for datacenters and makers of IT products alike, creating an ever more powerful on which to build compute, storage, and now networking or a mix of the three all in the same box. But that universal hardware substrate cuts both ways, and IT vendors have to be clever indeed if they hope to differentiate from their competitors.

So it is with the “Wolfcreek” storage platform from DataDirect Networks, which specializes in high-end storage arrays aimed at HPC, webscale, and high-end enterprise workloads. DDN started unveiling the Wolfcreek system last June

Scaling All Flash Arrays Up And Out was written by Timothy Prickett Morgan at The Next Platform.

First Burst Buffer Use at Scale Bolsters Application Performance

Over the last year, we have focused on the role burst buffer technology might play in bolstering the I/O capabilities on some of the world’s largest machines and have focused on use cases ranging from the initial target to more application-centric goals.

As we have described in discussions with the initial creator of the concept, Los Alamos National Lab’s, Gary Grider, the starting point for the technology was for moving the checkpoint and restart capabilities forward faster (detailed description of how this works here). However, as the concept developed over the years, some large supercomputing sites, including the National

First Burst Buffer Use at Scale Bolsters Application Performance was written by Nicole Hemsoth at The Next Platform.

Tesla Pushes Nvidia Deeper Into The Datacenter

If you are trying to figure out what impact the new “Pascal” family of GPUs is going to have on the business at Nvidia, just take a gander at the recent financial results for the datacenter division of the company. If Nvidia had not spent the better part of a decade building its Tesla compute business, it would be a little smaller and quite a bit less profitable.

In the company’s first quarter of fiscal 2017, which ended on May 1, Nvidia posted sales of $1.31 billion, up 13 percent from the year ago period, and net income hit $196

Tesla Pushes Nvidia Deeper Into The Datacenter was written by Timothy Prickett Morgan at The Next Platform.

Can Open Source Hardware Crack Semiconductor Industry Economics?

The running joke is that when a headline begs a question, the answer is, quite simply, “No.” However, when the question is multi-layered, wrought with dependencies that stretch across an entire supply chain, user bases, and device range, and across companies in the throes of their own economic and production uncertainties, a much more nuanced answer is required.

Although Moore’s Law is not technically dead yet, organizations from the IEEE to individual device makers are already thinking their way out of a box that has held the semiconductor industry neatly for decades. However, it turns out, that thought process is

Can Open Source Hardware Crack Semiconductor Industry Economics? was written by Nicole Hemsoth at The Next Platform.

IBM Research Lead Charts Scope of Watson AI Effort

Over the past few years, IBM has been devoting a great deal of corporate energy into developing Watson, the company’s Jeopardy-beating supercomputing platform. Watson represents a larger focus at IBM that integrates machine learning and data analytics technologies to bring cognitive computing capabilities to its customers.

To find out about how the company perceives its own invention, we asked IBM Fellow Dr. Alessandro Curioni to characterize Watson and how it has evolved into new application domains. Curioni, will be speaking on the subject at the upcoming ISC High Performance conference. He is an IBM Fellow, Vice President Europe and

IBM Research Lead Charts Scope of Watson AI Effort was written by Nicole Hemsoth at The Next Platform.

Shared Memory Pushes Wheat Genomics To Boost Crop Yields

Wheat has been an important part of the human diet for the past 9,000 years or so, and depending on the geography can comprise up to 40 percent to 50 percent of the diet within certain regions today.

But there is a problem. Pathogens and changing climate are adversely affecting wheat yields just as Earth’s population is growing, and the Genome Analysis Center (TGAC) is front and center in sequencing and assembling the wheat genome, a multi-year effort that is going to be substantially accelerated by some hardware and updated software.

With the world’s population expected to hit 10 billion

Shared Memory Pushes Wheat Genomics To Boost Crop Yields was written by Timothy Prickett Morgan at The Next Platform.

Facebook Flow Is An AI Factory Of The Future

We have been convinced for many years that machine learning, the kind of artificial intelligence that actually works in practice, not in theory, would be a key element of the next platform. In fact, it might be the most important part of the stack. And therefore, those who control how we deploy machine learning will, to a large extent, control the nature of future applications and the systems that run them.

Machine learning is the killer app for the hyperscalers, just like modeling and simulation were for supercomputing centers decades ago, and we believe we are only seeing the tip

Facebook Flow Is An AI Factory Of The Future was written by Timothy Prickett Morgan at The Next Platform.

Intel Stretches Deep Learning on Scalable System Framework

The strong interest in deep learning neural networks lies in the ability of neural networks to solve complex pattern recognition tasks – sometimes better than humans. Once trained, these machine learning solutions can run very quickly – even in real-time – and very efficiently on low-power mobile devices and in the datacenter.

However training a machine learning algorithm to accurately solve complex problems requires large amounts of data that greatly increases the computational workload. Scalable distributed parallel computing using a high-performance communications fabric is an essential part of what makes the training of deep learning on large complex datasets

Intel Stretches Deep Learning on Scalable System Framework was written by Nicole Hemsoth at The Next Platform.

Google And Friends Add Prometheus To Kubernetes Platform

There are a lot of moving parts in a modern platform, and in this regard, they are no different from the platforms made a generation earlier. But a modern platform has a lot more automation and is handling more dynamic workloads that are popping into and out of existence on different parts of a cluster like quantum particles, and it takes a higher level of sophistication to monitor and manage the stack and the apps running on it.

Frustration with existing open source monitoring tools like Nagios and Ganglia is why the hyperscaler giants created their own tools – Google

Google And Friends Add Prometheus To Kubernetes Platform was written by Timothy Prickett Morgan at The Next Platform.

Next Generation Supercomputing Strikes Data, Compute Balance

Although the future of exascale computing might be garnering the most deadlines in high performance computing, one of the most important stories unfolding in the supercomputing space, at least from a system design angle, is the merging of compute and data-intensive machines.

In many ways, merging both the compute horsepower of today’s top systems with the data-intensive support in terms of data movement, storage, and software is directly at odds with current visions of exascale supercomputers. Hence there appear to be two camps forming on either side of the Top 500 level centers; one that argues strongly in favor of

Next Generation Supercomputing Strikes Data, Compute Balance was written by Nicole Hemsoth at The Next Platform.

More than Moore: IEEE Set to Standardize on Uncertainty

Several decades ago, Gordon Moore made it far simpler to create technology roadmaps along the lines of processor capabilities, but as his namesake law begins to slow on the rails, the IEEE is stepping in to create a new, albeit more diverse roadmap for future systems.

The organization has launched a new effort to identify and trace the course of what follows Moore’s Law with the International Roadmap for Devices and Systems (IRDS), which will take a workload focused view of the mixed landscape and the systems that will be required. In other words, instead of pegging a

More than Moore: IEEE Set to Standardize on Uncertainty was written by Nicole Hemsoth at The Next Platform.

EMC Shoots For Explosive Performance With Isilon Nitro

Storage giant EMC, soon to be part of the Dell Technologies conglomerate, declared that this would be the year of all flash for the company when it launched its DSSD D5 arrays back in February. It was not kidding, and as a surprise at this weeks EMC World 2016 conference, the company gave a sneak peek at a future all-flash version of its Isilon storage arrays, which are also aimed at high performance jobs but which are designed to scale capacity well beyond that of the DSSD.

The DSSD D5 is an impressive beast, packing 100 TB of usable

EMC Shoots For Explosive Performance With Isilon Nitro was written by Timothy Prickett Morgan at The Next Platform.

Cloud Foundry Is Crossing The Chasm

IT managers at the world’s largest organizations have a lot of reasons to envy hyperscalers, including the fact that they seem to be flush with cash and it looks like they can buy or build just about anything their hearts desire.

While hyperscalers have to cope with scale issues, they do not have as much complexity, so they can pick a technology and run with it. Enterprises, on the other hand, are merging and acquiring all the time and have lots of silos of existing applications that cannot be thrown away.

The need to support existing as well as new

Cloud Foundry Is Crossing The Chasm was written by Timothy Prickett Morgan at The Next Platform.

The Long Future Ahead For Intel Xeon Processors

The personal computer has been the driver of innovation in the IT sector in a lot of ways for the past three and a half decades, but perhaps one of the most important aspects of the PC business is that it gave chip maker Intel a means of perfecting each successive manufacturing technology at high volume before moving it over to more complex server processors that would otherwise have lower yields and be more costly if they were the only chips Intel made with each process.

That PC volume is what gave Intel datacenter prowess, in essence, and it is

The Long Future Ahead For Intel Xeon Processors was written by Timothy Prickett Morgan at The Next Platform.

Mashing Up OpenStack With Hyperconverged Storage

While innovators in the HPC and hyperscale arenas usually have the talent and often have the desire to get into the code for the tools that they use to create their infrastructure, most enterprises want their software with a bit more fit and finish, and if they can get it so it is easy to operate and yet still in some ways open, they are willing to pay a decent amount of cash to get commercial-grade support.

OpenStack has pretty much vanquished Eucalyptus, CloudStack, and a few other open source alternatives from the corporate datacenter, and it is giving

Mashing Up OpenStack With Hyperconverged Storage was written by Timothy Prickett Morgan at The Next Platform.