FutureGrid Design and Implementation of a National Grid Test-Bed David Hancock – HPC Manager - Indiana University Hardware & Network Lead - FutureGrid

October 12, 2015Research Technologies IU in a nutshell $1.7B Annual Budget, >$100M annual IT budget Recent credit upgrade to AAA One university with – 8 campuses – 107,000 students – 3,900 faculty Nation’s 2nd largest school of medicine Serious HPC since 1990’s Research staff increased from since /50 Split in base and grant funding Large scale projects: TeraGrid, Open Science Grid (ATLAS Tier2 center), PolarGrid, Data Capacitor New Data Center opened in 2009

NSF Track Overview Track 1 – NCSA Blue Waters Track 2a – TACC Ranger Track2b – NICS Kraken Track 2d – Data Intensive High Performance System (SDSC) – Experimental High Performance System (GaTech) – Experimental High Performance Test-Bed (IU)

FutureGridFutureGrid The goal of FutureGrid is to support the research on the future of distributed, grid, and cloud computing. FutureGrid will build a robustly managed simulation environment and test-bed to support the development and early use in science of new technologies at all levels of the software stack: from networking to middleware to scientific applications. The environment will mimic TeraGrid and/or general parallel and distributed systems – FutureGrid is part of TeraGrid and one of two experimental TeraGrid systems (other is GPU) This test-bed will succeed if it enables major advances in science and engineering through collaborative development of science applications and related software. FutureGrid is a (small 5400 core) Science/Computer Science Cloud but it is more accurately a virtual machine based simulation environment

FutureGrid Partners Indiana University (Architecture, core software, Support) Purdue University (HTC Hardware) San Diego Supercomputer Center at University of California San Diego (INCA, Performance Monitoring) University of Chicago/Argonne National Labs (Nimbus) University of Florida (ViNe, Education and Outreach) University of Southern California Information Sciences Institute (Pegasus to manage experiments) University of Tennessee Knoxville (Benchmarking) University of Texas at Austin/Texas Advanced Computing Center (Portal) University of Virginia (OGF, User Advisory Board) Center for Information Services and GWT-TUD from Technische Universtität Dresden. (VAMPIR) Blue institutions host FutureGrid hardware

Other Important Collaborators Other Important Collaborators Early users from an application and computer science perspective and from both research and education Grid5000 and D-Grid in Europe Commercial partners such as – Eucalyptus …. – Microsoft (Dryad + Azure) – Application partners NSF TeraGrid – Tutorial at TG10 Open Grid Forum – Possible BoF Possibly Open Nebula, Open Cirrus Testbed, Open Cloud Consortium, Cloud Computing Interoperability Forum. IBM-Google-NSF Cloud, and other DoE/NSF/… clouds

FutureGrid Timeline October 2009 – Project Starts November 2009 – SC09 Demo January 2010 – Significant Hardware installed April 2010 – First Early Users May 2010 – FutureGrid network complete August 2010 – FutureGrid Annual Meeting September 2010 – All hardware, except shared memory system, available October 2011 – FutureGrid allocatable via TeraGrid process – first two years by user/science board

FutureGrid Usage Scenarios Developers of end-user applications who want to create new applications in cloud or grid environments, including analogs of commercial cloud environments such as Amazon or Google. – Is a Science Cloud for me? Is my application secure? Developers of end-user applications who want to experiment with multiple hardware environments. Grid/Cloud middleware developers who want to evaluate new versions of middleware or new systems. Networking researchers who want to test and compare different networking solutions in support of grid and cloud applications and middleware. Education as well as research Interest in performance testing requires that bare metal images are important

FutureGrid Hardware

Compute Hardware System type# CPUs# CoresTFLOPS Total RAM (GB) Secondary Storage (TB) Site Status Dynamically configurable systems IBM iDataPlex *IU New System Dell PowerEdge TACC New System IBM iDataPlex UC New System IBM iDataPlex SDSC Existing System Subtotal Systems not dynamically configurable Cray XT5m *IU New System Shared memory system TBD *IU New System 4Q2010 IBM iDataPlex UF New System High Throughput Cluster PU Existing System Subtotal Total

Storage Hardware System TypeCapacity (TB)File SystemSiteStatus DDN 9550 (Data Capacitor) 339LustreIUExisting System DDN GPFSUCNew System SunFire x417096ZFSSDSCNew System Dell MD300030NFSTACCNew System FutureGrid has a dedicated network (except to TACC) and a network fault and delay generator Experiments can be isolated by request Additional partner machines may run FutureGrid software and be supported (but allocated in specialized ways)

System Milestones New Cray System (xray) – Delivery: January 2010 – Acceptance: February 2010 – Available for Use: April 2010 New IBM Systems (india) – Delivery: January 2010 – Acceptance: March 2010 – Available for Use: May 2010 Dell System (tango) – Delivery: April 2010 – Acceptance: June 2010 – Available for Use: July 2010 Existing IU iDataPlex (sierra) – Move to SDSC: January 2010 – Available for Use: April 2010 Storage Systems (Sun & DDN) – Delivery: December 2009 – Acceptance: January 2010

Logical Diagram

Network Impairments Device Spirent XGEM Network Impairments Simulator for jitter, errors, delay, etc Full Bidirectional 10G w/64 byte packets up to 15 seconds introduced delay (in 16ns increments) 0-100% introduced packet loss in.0001% increments Packet manipulation in first 2000 bytes up to 16k frame size TCL for scripting, HTML for manual configuration

Network Milestones December 2009 – Setup and configuration of core equipment at IU – Juniper EX 8208 – Spirent XGEM January 2010 – Core equipment relocated to Chicago – IP addressing & AS # February 2010 – Coordination with local networks – First Circuits to Chicago Active March 2010 – Peering with TeraGrid & Internet2 April 2010 – NLR Circuit to UFL (via FLR) Active May 2010 – NLR Circuit to SDSC (via CENIC) Active

Global NOC Background ~65 total staff Service Desk: proactive & reactive monitoring 24x7x365, coordination of support Engineering: All operational troubleshooting Planning/Senior Engineering: Senior Network engineers dedicated to single projects Tool Developers: Developers of GlobalNOC tool suite

17 Supported Projects OmniPoP REN- ISAC

FutureGrid Architecture Open Architecture allows to configure resources based on images Managed images allows to create similar experiment environments Experiment management allows reproducible activities Through our modular design we allow different clouds and images to be “rained” upon hardware. Will support deployment of preconfigured middleware including TeraGrid stack, Condor, BOINC, gLite, Unicore, Genesis II

Software Goals Open-source, integrated suite of software to – instantiate and execute grid and cloud experiments. – perform an experiment – collect the results – tools for instantiating a test environment TORQUE, Moab, xCAT, bcfg, and Pegasus, Inca, ViNE, a number of other tools from our partners and the open source community Portal to interact – Benchmarking 10/12/2015http://futuregrid.org19

Draft GUI for FutureGrid Dynamic Provisioning Draft GUI for FutureGrid Dynamic Provisioning

Command line fg-deploy-image – host name – image name – start time – end time – label name fg-add – label name – framework hadoop – version 1.0 Deploys an image on a host Adds a feature to a deployed image 10/12/2015http://futuregrid.org21

FG Stratosphere Objective – Higher than a particular cloud – Provides all mechanisms to provision a cloud on a given FG hardware – Allows the management of reproducible experiments – Allows monitoring of the environment and the results Risks – Lots of software – Possible multiple path to do the same thing Good news – We worked in a team, know about different solutions and have identified a very good plan – We can componentize Stratosphere 10/12/2015http://futuregrid.org22

Change underlying system to support current user demands Linux, Windows, Xen/KVM, Nimbus, Eucalyptus Stateless images Shorter boot times Easier to maintain Stateful installs Windows Use Moab to trigger changes and xCAT to manage installs 10/12/201523http://futuregrid.org Dynamic Provisioning

xCAT and Moab xCAT uses installation infrastructure to perform installs creates stateless Linux images changes the boot configuration of the nodes remote power control and console Moab meta-schedules over resource managers  TORQUE and Windows HPC control nodes through xCAT  changing the OS 10/12/201524http://futuregrid.org

Experiment Manager Objective – Manage the provisioning for reproducible experiments – Coordinate workflow of experiments – Share workflow and experiment images – Minimize space through reuse Risk – Images are large – Users have different requirements and need different images 10/12/2015http://futuregrid.org25

AcknowledgementsAcknowledgements FutureGrid NSF Award OCI NSF Solicitation – ViNe Nimbus Eucalyptus VAMPIR Pegasus -