• Lessons Learned From Facebook’s Split Network Backbone

  May 2, 2017 Timothy Prickett Morgan

  

  Distributed applications, whether they are containerized or not, have a lot of benefits when it comes to modularity and scale. But in a world of feature creep on all applications, whether they are internally facing ones running a business or hyperscale consumer applications like Google’s search engine or Facebook’s social media network, these distributed applications put a huge strain on the network.

  This, more than any other factor, is why network costs are rising faster than any other aspect of the datacenter. Gone are the days when everything was done in three or four tiers, with a Web server like …

  Read more

• Machine Learning Gets An InfiniBand Boost With Caffe2

  April 19, 2017 Timothy Prickett Morgan

  

  Scaling the performance of machine learning frameworks so they can train larger neural networks – or so the same training a lot faster – has meant that the hyperscalers of the world who are essentially creating this technology have had to rely on increasingly beefy compute nodes, these days almost universally augmented with GPUs.

  There is a healthy rivalry between the hyperscalers over who has the best machine learning framework and the co-designed iron to take the best advantage of its capabilities. At its F8 developer conference, Facebook not only rolled out a significantly tweaked variant of the open source …

  Read more

• A Look at Facebook’s Interactive Neural Network Visualization System

  April 12, 2017 Nicole Hemsoth

  

  There has been much discussion about the “black box” problem of neural networks. Sophisticated models can perform well on predictive workloads, but when it comes to backtracking how the system came to its end result, there is no clear way to understand what went right or wrong—or how the model turned on itself to arrive a conclusion.

  For old-school machine learning models, this was not quite the problem it is now with non-linear, hidden data structures and countless parameters. For researchers deploying neural networks for scientific applications, this lack of reproducibility from the black box presents validation hurdles, but for …

  Read more

• Facebook Pushes The Search Envelope With GPUs

  March 27, 2017 Timothy Prickett Morgan

  

  An increasing amount of the world’s data is encapsulated in images and video and by its very nature it is difficult and extremely compute intensive to do any kind of index and search against this data compared to the relative ease with which we can do so with the textual information that heretofore has dominated both our corporate and consumer lives.

  Initially, we had to index images by hand and it is with these datasets that the hyperscalers pushed the envelope with their image recognition algorithms, evolving neural network software on CPUs and radically improving it with a jump to …

  Read more

• Open Hardware Pushes GPU Computing Envelope

  March 17, 2017 Timothy Prickett Morgan

  

  The hyperscalers of the world are increasingly dependent on machine learning algorithms for providing a significant part of the user experience and operations of their massive applications, so it is not much of a surprise that they are also pushing the envelope on machine learning frameworks and systems that are used to deploy those frameworks. Facebook and Microsoft were showing off their latest hybrid CPU-GPU designs at the Open Compute Summit, and they provide some insight into how to best leverage Nvidia’s latest “Pascal” Tesla accelerators.

  Not coincidentally, the specialized systems that have been created for supporting machine learning workloads …

  Read more

• A Peek Inside Facebook’s Server Fleet Upgrade

  March 13, 2017 Timothy Prickett Morgan

  

  Having a proliferation of server makes and models over a span of years in the datacenter is not a huge deal for most enterprises. They cope with the diversity because they support a diversity of application and can kind of keep things isolated and, moreover, IT may be integral to their product or service, but it is usually not the actual product or service that they sell.

  Not so with hyperscalers and cloud builders. For them, the IT is the product, and keeping things as monolithic and consistent as possible lowers the cost of goods purchased through higher volumes and …

  Read more

• Microsoft Azure Goes Back To Rack Servers With Project Olympus

  November 1, 2016 Timothy Prickett Morgan

  

  The thing we hear time and time again from the hyperscalers is that technology is a differentiator, but supply chain can make or break them. Designing servers, storage, switching, and datacenters is fun, but if all of the pieces can’t be brought together at volume, and at a price that is the best in the industry, then their operations can’t scale.

  It is with this in mind that we ponder Microsoft’s new “Project Olympus” hyperscale servers, which it debuted today at the Zettastructure conference in London. Or, to be more precise, the hyperscale server designs that it has created but …

  Read more

• One Rack To Stack Them All

  August 8, 2016 Timothy Prickett Morgan

  

  Stacking up electronics equipment in precise form factors that slide into standard racks is not a new idea, and in fact it is one that predates the modern era of computing. As is the case with any standard, the constraints it imposes brings order to the market while at the same time restricting it, and making any substantial change in something as fundamental as the datacenter rack requires a pretty significant payback.

  Any standard also requires volume manufacturing to really take off and yield benefits, and this has certainly not happened with rack-scale architectures to date. The time is perhaps …

  Read more

• Facebook Flow Is An AI Factory Of The Future

  May 10, 2016 Timothy Prickett Morgan

  

  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 …

  Read more

• Datacenters To Get A High Fiber Bandwidth Diet

  March 18, 2016 Timothy Prickett Morgan

  

  Might doesn’t make right, but it sure does help. The hyperscalers, cloud builders, and co-location datacenters of the world that operate at massive scale have an interconnectivity problem, and it has nothing to do with the feeds and speeds of a switch. What is choking them is the Ethernet cable, and they are working together to do something about it.

  It may be hard to believe, but the cabling in the datacenter that allows for the communication between servers, storage, and the outside world in a modern distributed application cost more and are more of a limiting factor in a …

  Read more

Previous Articles