Interactive Launch of 16,000 Microsoft Windows Instances on a Supercomputer

Simulation, machine learning, and data analysis require a wide range of software which can be dependent upon specific operating systems, such as Microsoft Windows. Running this software interactively on massively parallel supercomputers can present many challenges. Traditional methods of scaling Microsoft Windows applications to run on thousands of processors have typically relied on heavyweight virtual machines that can be inefficient and slow to launch on modern manycore processors. This paper describes a unique approach using the Lincoln Laboratory LLMapReduce technology in combination with the Wine Windows compatibility layer to rapidly and simultaneously launch and run Microsoft Windows applications on thousands of cores on a supercomputer. Specifically, this work demonstrates launching 16,000 Microsoft Windows applications in 5 minutes running on 16,000 processor cores. This capability significantly broadens the range of applications that can be run at large scale on a supercomputer.

[1]  Hyung Seok Kim,et al.  Interactive Grid Computing at Lincoln Laboratory , 2006 .

[2]  Ming Mao,et al.  A Performance Study on the VM Startup Time in the Cloud , 2012, 2012 IEEE Fifth International Conference on Cloud Computing.

[3]  Jeremy Kepner,et al.  D4M: Bringing associative arrays to database engines , 2015, 2015 IEEE High Performance Extreme Computing Conference (HPEC).

[4]  Robert N. M. Watson,et al.  Jails: confining the omnipotent root , 2000 .

[5]  Marty Humphrey,et al.  Assessing the Value of Cloudbursting: A Case Study of Satellite Image Processing on Windows Azure , 2011, 2011 IEEE Seventh International Conference on eScience.

[6]  Jeremy Kepner,et al.  Scalability of VM provisioning systems , 2016, 2016 IEEE High Performance Extreme Computing Conference (HPEC).

[7]  Jeremy Kepner,et al.  LLSuperCloud: Sharing HPC systems for diverse rapid prototyping , 2013, 2013 IEEE High Performance Extreme Computing Conference (HPEC).

[8]  Greg Yeric,et al.  Moore's law at 50: Are we planning for retirement? , 2015, 2015 IEEE International Electron Devices Meeting (IEDM).

[9]  Jeremy Kepner,et al.  Interactive Supercomputing on 40,000 Cores for Machine Learning and Data Analysis , 2018, 2018 IEEE High Performance extreme Computing Conference (HPEC).

[10]  Robert K. Cunningham,et al.  Computing on masked data: a high performance method for improving big data veracity , 2014, 2014 IEEE High Performance Extreme Computing Conference (HPEC).

[11]  Carl A. Waldspurger,et al.  Memory resource management in VMware ESX server , 2002, OSDI '02.

[12]  Al Geist,et al.  A survey of high-performance computing scaling challenges , 2017, Int. J. High Perform. Comput. Appl..

[13]  David A. Lifka.  High performance computing with microsoft windows 2000 , 2001, Proceedings 42nd IEEE Symposium on Foundations of Computer Science.

[14]  Peng Xu,et al.  Wine , 2006, A Handbook of Food Processing in Classical Rome.

[15]  Sanjay Ghemawat,et al.  MapReduce: Simplified Data Processing on Large Clusters , 2004, OSDI.

[16]  Jeff Dike,et al.  User-mode Linux , 2006, Annual Linux Showcase & Conference.

[17]  Jeremy Kepner,et al.  LLgrid: Enabling On-Demand Grid Computing with gridMatlab and pMatlab , 2004 .

[18]  Jeremy Kepner,et al.  Big Data strategies for Data Center Infrastructure management using a 3D gaming platform , 2015, 2015 IEEE High Performance Extreme Computing Conference (HPEC).

[19]  Jeremy Kepner,et al.  Driving big data with big compute , 2012, 2012 IEEE Conference on High Performance Extreme Computing.

[20]  Thomas N. Theis,et al.  The End of Moore's Law: A New Beginning for Information Technology , 2017, Computing in Science & Engineering.

[21]  Jeremy Kepner,et al.  Scalable System Scheduling for HPC and Big Data , 2017, J. Parallel Distributed Comput..

[22]  Dirk Merkel,et al.  Docker: lightweight Linux containers for consistent development and deployment , 2014 .

[23]  Jeremy Kepner,et al.  Achieving 100,000,000 database inserts per second using Accumulo and D4M , 2014, 2014 IEEE High Performance Extreme Computing Conference (HPEC).

[24]  Kirill Kolyshkin,et al.  VIRTUALIZATION IN LINUX , 2006 .

[25]  Linus Torvalds,et al.  Linux : a Portable Operating System , 2011 .

[26]  Jeremy Kepner,et al.  LLMapReduce: Multi-level map-reduce for high performance data analysis , 2016, 2016 IEEE High Performance Extreme Computing Conference (HPEC).

[27]  Jeremy Kepner,et al.  Enabling on-demand database computing with MIT SuperCloud database management system , 2015, 2015 IEEE High Performance Extreme Computing Conference (HPEC).

[28]  Andy B. Yoo,et al.  Approved for Public Release; Further Dissemination Unlimited X-ray Pulse Compression Using Strained Crystals X-ray Pulse Compression Using Strained Crystals , 2002 .

[29]  Jeremy Kepner,et al.  Scheduler technologies in support of high performance data analysis , 2016, 2016 IEEE High Performance Extreme Computing Conference (HPEC).

[30]  Kenji Takeda,et al.  HPC On Dec Alphas And Windows NT , 1999 .

[31]  Jeremy Kepner,et al.  Dynamic distributed dimensional data model (D4M) database and computation system , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[32]  Peter Braam,et al.  The Lustre Storage Architecture , 2019, ArXiv.

[33]  Jeremy Kepner,et al.  D4M 2.0 schema: A general purpose high performance schema for the Accumulo database , 2013, 2013 IEEE High Performance Extreme Computing Conference (HPEC).

[34]  Ramakrishnan Rajamony,et al.  An updated performance comparison of virtual machines and Linux containers , 2015, 2015 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS).

[35]  Yingjie Xia,et al.  Parallel geospatial analysis on windows HPC platform , 2010, 2010 The 2nd Conference on Environmental Science and Information Application Technology.

[36]  John Evans,et al.  HPC in a HEP lab: lessons learned from setting up cost-effective HPC clusters , 2015 .