Bringing the Arm ecosystem together

Abstract

The next decade promises the democratization of connectivity to every device. Significant drops in the cost of connectivity mean that every form of electrical device—every child’s toy, every household’s appliances, and every sensor—will become connected to the Internet. Tens of billions of these devices are controlled by microcontrollers, a class of device particularly ill-prepared for the security challenges of internet connectivity. What is required to secure these billions of devices? Microsoft defines 7 properties that define the minimum standard for any device connecting to the internet and in this talk we’ll dive into each as we make the case for a better standard for security and connectivity.

Ed Nightingale

Partner Director of Engineering at Microsoft Azure Sphere
Ed Nightingale has been part of Azure Sphere from its inception and has filled many different roles as the product has grown from research to product. Currently, Ed is the Partner Director of Engineering for Azure Sphere where he is responsible for the overall software and hardware engineering effort on the team. Prior to Azure Sphere, Ed was a systems researcher and software engineer. Ed has authored papers in top systems conferences such as OSDI and SOSP. He has co-authored 20 conference and journal publications, has won 6 best paper awards, and even helped to set the world record in disk-to-disk sorting (http://sortbenchmark.org). He has also worked as an engineer and as an engineering manager running a large-scale distributed storage service. Ed really enjoys building operating systems and large-scale distributed systems.

Abstract

SOC vendors are often regarded as not being good opensource citizens. They tend to work downstream with upstream being an afterthought -- if it's even being thought of at all. This is contrary to opensource community practices and can cause friction when engineers operating in these different modes try to interact. There are often good reasons, and sometimes not-so-good reasons, for this type of behavior. In this talk I will attempt to give more insight into why SOC vendors are downstream-focused, what types of issues motivate this behavior, and how we can better bridge these ecosystems to "pop the bubble".

Bryan Huntsman

Senior Director at Qualcomm
Bryan Huntsman is a Senior Director at the Qualcomm Innovation Center and is responsible for Linux Kernel development on Qualcomm Snapdragon processors for mobile and IoT markets. Over the past nine years, he has set up a team of engineers and engineering processes that contributed to the launch of over a billion Android devices. Prior to joining Qualcomm, Bryan spent 10 years in the telecommunications industry working as a SW/FW developer for embedded, real-time systems. Bryan has a B.S in Computer Engineering from the University of Florida.

Abstract

The University of Michigan(U-M), in partnership with Marvell, has built a Data Analytics cluster with ThunderX processors based on the Hadoop and Spark family of tools. We present information on the deployment and use of the cluster, which uses software packaged by the Linaro Big Data teams. We also present feedback to the Big Data on ARM community and future plans for development for Big Data on ARM at U-M as well.

Jeremy Hallum

Research Computing Manger, Advanced Research Computing - Technology Services, University of Michigan
Jeremy Hallum is the manager responsible for Storage, Private Cloud, and Data Science Operations teams at ARC-TS at the University of Michigan. His teams are responsible for the deployment of the storage systems, Yottabyte Research Cloud, and Marvell data science Cluster. Jeremy earned his undergraduate degree from Minnesota State University in Astronomy and Mathematics in 1996, and a Masters degree from Boston University in 1999 in Astronomy. His interests lie in developing usability for advanced computing resources, user training and secure use of advanced computing resources.

Abstract

Fugaku is the flagship next generation national supercomputer being developed by Riken R-CCS and Fujitsu in collaboration. Fugaku will have hyperscale datacenter class resource in a single exascale machine, with more than 150,000 nodes of sever-class Fujitsu A64fx many-core Arm CPUs with the new SVE (Scalable Vector Extension) with low precision math for the first time in the world, accelerating both HPC and AI workloads, augmented with HBM2 memory paired with each CPU, exhibiting nearly a Terabyte/s memory bandwidth for both HPC and AI rapid data movements. A64fx has demonstrated up to 8 times the performance of mainstream "Platinum" server processors for HPC workloads, and this game-changing performance has been realized through extensive co-design process involving the entire Japanese HPC community resulting in a highly HPC-optimzed CPU architecture. Fugaku also will likely to be the premier big data and AI/ML infrastructure; currently, we are conducting research to scale deep learning to more than 100,000 nodes on Fugaku, where we expect to obtain near top GPU-class performance on each node. The talk will give an overview of A64fx/Fugaku as well as cover some of the software ecosystem efforts to enable high-end HPC / BD / AI.

Satoshi Matsuoka

Director, RIKEN Center for Computational Science
Satoshi Matsuoka had been a Full Professor at the Global Scientific Information and Computing Center (GSIC), the Tokyo Institute of Technology since 2000, and the director of the joint AIST-Tokyo Tech. Real World Big Data Computing Open Innovation Laboratory (RWBC-OIL) since 2017, and will become a Specially Appointed Professor at Tokyo Tech starting 2018 along with his directorship at R-CCS. He received his Ph. D. from the University of Tokyo in 1993. He has been the leader of the TSUBAME series of supercomputers that have won many accolades such as world #1 in power-efficient computing. He also leads various major supercomputing research projects in areas such as parallel algorithms and programming, resilience, green computing, and convergence of big data/AI with HPC. He has written over 500 articles according to Google Scholar, and chaired numerous ACM/IEEE conferences, including the Program Chair at the ACM/IEEE Supercomputing Conference (SC13) in 2013. He is a Fellow of the ACM and European ISC, and has won many awards, including the JSPS Prize from the Japan Society for Promotion of Science in 2006, presented by his Highness Prince Akishino; the ACM Gordon Bell Prize in 2011; the Commendation for Science and Technology by the Minister of Education, Culture, Sports, Science and Technology in 2012; the 2014 IEEE-CS Sidney Fernbach Memorial Award, the highest prestige in the field of HPC; and recently HPDC 2018 Achievement Award from ACM.

Abstract

Over the last 10+ years, we have been developing the Coccinelle C-program matching and transformation tool. Coccinelle allows developers to match and transform C code via a generalization of the familiar patch syntax. As Coccinelle is independent of the C preprocessor, Coccinelle rules can be applied across a code base, regardless of configuration constraints. Over the years, Coccinelle has been extensively used in Linux kernel development, resulting in over 7000 commits to the Linux kernel, and has found its place as part of the Linux kernel development process. This talk gives an introduction to the use of Coccinelle and gives an overview of its impact on the Linux kernel.

Julia Lawall

Senior Research Scientist at Inria
Julia Lawall is a Senior Research Scientist at Inria. Her research is at the intersection of programming languages and operating systems. She develops the tool Coccinelle and has over 2000 patches in the Linux kernel based on this work.

Abstract

Arm-based systems are showing up in the HPC community and new silicon is coming. The architecture has also been selected for several of the exascale projects worldwide. Brent will talk about the aspects of Arm that are attractive to the HPC community, updates on projects and what we as a community can do to help accelerate adoption in this space.

Brent Gorda

Sr. Director, Worldwide HPC at Arm
Brent has a long history of working in supercomputing community. Starting in the mid-80’s in Canada, he wrote compilers for Myrias Research. In the early 90’s he moved to the Lawrence Livermore National Laboratory and worked on the adoption of parallel computing. In the early to mid 2000’s, he ran the BlueGene project for Livermore as well as the Lustre file system contract. Brent founded and was CEO at Whamcloud, a company formed to keep Lustre in play for HPC. After selling Whamcloud to Intel, he was General Manager for the High Performance Data Division for 5+ years. Along the way, Brent had a hand in starting Software-Carpentry.com, building the first Java compiler and founding the Student Cluster Competition now held worldwide at HPC events.