• Rescale Night: The Future of Cloud HPC Simulation

    Microsoft, San Francisco Office

    You are cordially invited to our inaugural Rescale Night in San Francisco on Thursday, January 26, 2017. Join us for an evening of lively discussion about the Future of Cloud HPC Simulation. Hear from Boom Technology, Microsoft, ANSYS, and more! Register at www.rescale.com/events

  • Live Migration of HPC workloads with Terminal.com

    Title: "Live Migration of HPC workloads with Terminal.com" Presenter: Joshua Goldbard, VP, Terminal.com Short bio: Joshua Goldbard is a VP at Terminal.com working on next-generation virtualization technology. Prior to his time at Terminal, he spent a large amount of time in executive positions at Telecommunications and computing companies. Joshua enjoys craft beer, infrastructure of all kinds and loud music. Abstract: This talk will discuss Terminal's work in live migration of Linux Workloads (of an HPC and non-HPC nature). Specifically, we'll cover what Terminal does, the modifications we've made to low-level code and how those changes benefit our users. Of note, we'll discuss Terminal's snapshotting capabilities and the live migration feature called "teleport". The presentation will include a live demo, an architecture discussion and some utilization tips. Walk-in: 6.30 Presentation:[masked] Wrap-up:[masked]

    1
  • Rescaling the Future of HPC and Simulation

    Geekdom SF

    Title: "Rescaling the Future of HPC and Simulation" Presenter: Joris Poort, CEO and Founder of Rescale Short bio: Prior to founding Rescale, Joris worked as an associate at McKinsey & Company consulting on product development engagements in the high-tech sector. Joris began his career at Boeing, where he worked for four years as a structural and software engineer on the 787 program, optimizing the design of the tail and wings. Joris holds an M.B.A. with distinction from Harvard Business School, an M.S. in Aeronautics and Astronautics magna cum laude from the University of Washington, and a B.S. in Mechanical Engineering and minor in Applied Mathematics magna cum laude from the University of Michigan Abstract: Rescale’s cloud simulation platform is the world’s leading solution for the secure deployment of simulation software and high performance computing (HPC) solutions in the enterprise. The platform is deployed securely and seamlessly to global teams via the web browser in an application environment built for engineers and scientists backed by the largest commercially available HPC backbone. Rescale’s customers enjoy a significant simulation advantage over their competition through the agility to seamlessly scale both software and hardware solutions based on internal demand. Users of the simulation platform experience numerous benefits, including, increased efficiency of IT resources and capex reductions, faster design cycles with dramatically improved time to market, and superior product designs through increased simulation accuracy and integration of simulation processes. Rescale is an enterprise software company funded by numerous top-tier venture investors including Jeff Bezos, Richard Branson, and Peter Thiel--the founders of Amazon, Virgin and Paypal, respectively. Rescale has a global set of partners and customers that utilizes Rescale's technology to improve how simulation is utilized for anything from Formula 1 cars to spacecraft to drug development with a customer base including Fortune Global 500 enterprises across numerous sectors from aerospace, automotive, life science to energy verticals.

    8
  • Usable Exascale and Beyond Moore's Law

    Geekdom SF

    [masked]: welcome reception [masked]: "Usable Exascale and Beyond Moore's Law" Horst Simon -- Deputy Laboratory Director @ Lawrence Berkeley National Laboratory Abstract As documented by the TOP500, high performance computing (HPC) has been the beneficiary of uninterrupted growth, and performance of the top HPC systems doubled about every year until 2004, when Dennard scaling tapered off. This was based on the contributions of Moore’s law and the increasing parallelism in the highest end system. Continued HPC system performance increases were then obtained by doubling parallelism. However, over the last five years HPC performance growth has been slowing measurably, and in this presentation several reasons for this slowdown will be analyzed. To reach usable exascale performance over the next decade, some fundamental changes will have to occur in HPC systems architecture. In particular, a transition from a compute centric to a data movement centric point of view needs to be considered. Alternatives including quantum and neuromorphic computing have also been considered. The prospects of these technologies for post-Moore’s Law supercomputing will be explored. Bio Horst Simon is an internationally recognized expert in computer science and applied mathematics and the Deputy Director of Lawrence Berkeley National Laboratory (Berkeley Lab). Simon joined Berkeley Lab in early 1996 as director of the newly formed National Energy Research Scientific Computing Center (NERSC), and was one of the key architects in establishing NERSC at its new location in Berkeley. Under his leadership NERSC enabled important discoveries for research in fields ranging from global climate modeling to astrophysics. Simon was also the founding director of Berkeley Lab's Computational Research Division, which conducts applied research and development in computer science, computational science, and applied mathematics. [masked]: wrap-up

    8
  • Planning into 2015

    Geekdom SF

    [masked]: welcome reception [masked]: Planning 2015 and 2-5 min self presentations from the meetup members [masked]: wrap-up

    1
  • ScaleMP: Virtualization for Flexible HPC Environments

    [masked]: welcome reception (Beer, Holiday Festivities) [masked]: "Virtualization for Flexible HPC Environments" Michael Schulman -- Director of Marketing @ ScaleMP High end virtualization for HPC environments is an extremely valuable technology which helps to reduce costs and create a more flexible HPC environment. By aggregating industry-standard systems into powerful, large scale SMPs, organizations can quickly react to various workloads. Applications can get full access to terabytes of memory and hundreds of compute cores in a simple and easy to use system. The latest technology is available to users and in addition they can get access to massive quantities of compute cores and memory. In addition to fully supported versions, a free version is available so that demanding users can experience how their applications can scale, with out limiting the apps to small amounts of memory. [masked]: (Optional) Lightning talks: 2-3 min self presentations from the meetup members

    1
  • Cluster scheduling with Mesos + NERSC web toolkit

    [masked]: welcome reception [masked]: "Introducing NEWT: An API and Framework for RESTful High Performance Computing" Shreyas Cholia -- Deputy Group Leader, Data & Analytics Group @ NERSC NEWT is bringing High Performance Computing (HPC) to the web through easy to write web applications. It is a web service that allows you to access computing resources at HPC centers like NERSC through a simple RESTful API. The NEWT API and web service will let you interact with NERSC through simple HTTP urls and commands. NEWT responds to client requests using JSON. This makes it very easy to build powerful web applications using nothing but HTML and Javascript.make HPC resources accessible and useful to scientists who are more comfortable with the web than they are with compilers, batch queues, or command-lines. [masked]: "Mesos: building blocks for advanced cluster scheduling" Niklas Nielsen -- Distributed Systems Engineer @ Mesosphere Apache Mesos is a cluster manager that provides efficient resource isolation and sharing across distributed applications. In this talk, Niklas will go over how Mesos works as a library for rapidly developing new and port existing schedulers for multi-tenant clusters.

    9
  • Four rules to govern system performance

    Geekdom SF

    7-7.30 : Welcome Reception [masked]: Udi Davidovich, Director of Platform Services @ Tidemark "Four rules to govern system performance" The performance of any system is governed by four fundamental rules that ties together concurrency, throughput, response time and utilization. After reviewing a formal model of a multi-tiered parallelized system, we will identify key measurable metrics, derive critical performance metrics, and investigate how the rules determine the overall performance of the system across a range of loads.

    5
  • Converging Interconnect Requirements for HPC and Warehouse Scale Computing

    [masked]: welcome reception [masked]: "Converging Interconnect Requirements for HPC and Warehouse Scale Computing" John Shalf -- Department Head for Computer Science for the Computing Research Division @ Lawrence Berkeley Lab -- Chief Technology Officer @ NERSC This talk will discuss the emerging trends towards convergence of cloud and HPC interconnect requirements due to both market forces and increasingly aligned performance requirements. One point of divergence between HPC and Cloud computing is the interconnect technology. Energy efficiency concerns and the cost effectiveness of using commodity technology has strongly aligned both HPC and cloud computing in terms of processor and cooling technology, but the interconnects have remained a big differentiating factor. HPC has typically required custom high performance interconnects for tightly-coupled inward-facing traffic. Cloud computing has tended to focus on standards-based Ethernet with TCP/IP for outward-facing wide area networking. However, with the rise of large scale data analytics applications, warehouse scale computing systems have observed that large fractions of their communications traffic is inward-facing, and that the performance of the inward-facing datacenter network has become a primary limiting factor for performance. Furthermore, recent acquisition of high performance interconnect technologies by Intel for integration directly onto the processor has fundamentally changed the ecosystem for commodity server chips. The technologies are converging, but will the ends meet? The talk will examine trends in both requirements for datacenter workloads as well as emerging technology trends and what it means for the future of HPC and cloud environments. [masked]: socializing

    5
  • Kickoff meeting

    Location visible to members

    A meeting to discuss recent updates in high-performance computing. Approximate agenda: [masked] : Welcome Reception [masked]: Talks: + HPC updates from NERSC@LBL / Yushu Yao + Notes about user experience in HPC: what we learn from Interviewing scientists / Miha Ahronovitz + Latest updates from The UberCloud - on-demand HPC in the cloud / Burak Yenier + Exabyte.io - web platform for scientific computing / Timur Bazhirov [masked]: Wrapping up, Q&A with speakers

    16