In this podcast, the Radio Free HPC team looks at why it’s so difficult for new processor architectures to gain traction in HPC and the datacenter. Plus, we introduce a new regular feature for our show: The Catch of the Week.
Intel and Hewlett Packard Enterprise (HPE) have recently created two new Centers of Excellence (CoE) to help customers gain hands-on experience with High Performance Computing (HPC). This plus collaboration with customers on implementing the latest technology solutions are highlights being celebrated by the two companies on the one-year anniversary of their alliance.
In this podcast, the Radio Free HPC team reviews the recent 2016 Intel Developer Forum. “How will Intel return to growth in the face of a declining PC market? At IDF, they put the spotlight on IoT and Machine Learning. With new threats rising from the likes of AMD and Nvidia, will Chipzilla make the right moves? Tune in to find out.”
In this video, D-Wave Systems Founder Eric Ladizinsky presents: The Coming Quantum Computing Revolution. “Despite the incredible power of today’s supercomputers, there are many complex computing problems that can’t be addressed by conventional systems. Our need to better understand everything, from the universe to our own DNA, leads us to seek new approaches to answer the most difficult questions. While we are only at the beginning of this journey, quantum computing has the potential to help solve some of the most complex technical, commercial, scientific, and national defense problems that organizations face.”
“Few fields are moving faster right now than deep learning,” writes Buck. “Today’s neural networks are 6x deeper and more powerful than just a few years ago. There are new techniques in multi-GPU scaling that offer even faster training performance. In addition, our architecture and software have improved neural network training time by over 10x in a year by moving from Kepler to Maxwell to today’s latest Pascal-based systems, like the DGX-1 with eight Tesla P100 GPUs. So it’s understandable that newcomers to the field may not be aware of all the developments that have been taking place in both hardware and software.”
In this podcast, the Radio Free HPC team looks HPE’s pending acquisition of SGI. “Will the acquisition be good for SGI and HP customers? Our RFHPC team is in unprecedented agreement that indeed it will. The key, however, to HPE’s success will be keeping the SGI people. Rich thinks this acquisition will potentially give HPE the engineering talent it needs to compete with Cray at the high end of the market.”
Is Machine Learning more of a Data Movement problem than a Processing problem? In this podcast, the Radio Free HPC team looks at use cases for Machine Learning where data locality is critical for performance. “Most of the Machine Learning hearing stories we hear involve a central data repository. Henry says he is not hearing enough about how Machine Learning is going to deal with the problem of massive data streams from things like sensors. Such data, he contends, will have to be processed at the source.”
“We have been working on developing a number of tools that enable users to quantify power and performance in both software and hardware, and then design a more efficient system. We can also utilize the tools to predict the performance of a piece of software on a system that may not be available or does not yet exist – the aim is to take the guesswork away from novel system design.”
Cray’s Steve Scott presented this talk at The Digital Future Conference. “Research and development at Cray is guided by our adaptive supercomputing vision. This vision is focused on delivering innovative, next-generation products that integrate diverse processing technologies into a unified architecture, enabling customers to surpass today’s limitations and meeting the market’s demand for realized performance.”
In this special guest feature, Rob Farber writes that a study done by Kyoto University Graduate School of Medicine shows that code modernization can help Intel Xeon processors outperform GPUs on machine learning code. “The Kyoto results demonstrate that modern multicore processing technology now matches or exceeds GPU machine-learning performance, but equivalently optimized software is required to perform a fair benchmark comparison. For historical reasons, many software packages like Theano lacked optimized multicore code as all the open source effort had been put into optimizing the GPU code paths.”