XSEDE Gateways at SC15

XSEDE Science Gateway and Scientific Workflow Office Hours

  • Summary: Drop by the XSEDE booth and talk to XSEDE experts on how to get help using science gateways and scientific workflows, finding a gateway that meets your needs, or building your own gateway.

  • Presenters: Marlon Pierce (10-11:00 am Tuesday & 10-11:30 Wednesday), Sudhakar Pamidighantam, Yan Liu (11:30-12 Tue & Wed), Suresh Marru (Tuesday 10 am - Noon, Wed - 10 am to 11.30am)

  • Location: XSEDE SC15 booth #1741

  • Time: Tuesday, 10 am - 12 noon; Wednesday, 10 am - 12 noon

UltraScan: An HPC science gateway for biophysics

  • Summary: The UltraScan Science Gateway facilitates the analysis of biophysical experiments. The Gateway is used to model data from analytical ultracentrifuges using a finite element simulator and various HPC-enabled optimization algorithms. The software uses a web-based LIMS system and Apache Airavata to submit supercomputing jobs to HPC infrastructure offered by XSEDE (US), PRACE (Europe), and various local clusters to perform the computationally demanding analysis. This talk presentation provides an overview of this architecture and show examples of the results.

  • Presenters: Borries Demeler

  • Location: IU Pervasive Technology Institute Booth 532

  • Time:  Tuesday, 1:00-1:30 pm

Apache Airavata: Enabling Science with Science Gateways

  • Summary: We provide an overview of science gateway-enabled scientific research enabled by collaborations between science gateway developers, researchers, and the IU Science Gateway Group using Apache Airavata middleware. Highlights include applications in computational chemistry, biophysics, bioinformatics, engineering, and computer science.

  • Presenters: Marlon Pierce, Sudhakar Pamidighantam, Suresh Marru

  • Location: IU Pervasive Technology Institute Booth 532

  • Time: Monday, 8-8:30 pm

Building Science Gateways with Apache Airavata and SciGaP

  • Summary: Science gateways provide science-centric views of cyberinfrastructure. As the popularity of science gateways grows and cloud computing technologies proliferate, it is necessary for science gateway providers to adopt DevOps strategies for deploying and operating both their gateways and the science applications that they support on clouds.This presentation shows how Apache Airavata, a multi-tenanted science gateway framework, can be elastically and fault-tolerantly operated on Indiana University’s OpenStack-based Jetstream cloud computing resource using Apache Mesos and Marathon to manage Docker-wrapped applications. It will also show how these same tools can be used to manage the scheduling of Dockerized scientific applications via Apache Airavata, providing a lightweight scheduling solution for gateways using cloud computing as their back-end computing engine.

  • Presenters: Madhu Govindaraju, Suresh Marru, Marlon Pierce

  • Location: IU Pervasive Technology Institute Booth 532

  • Time: Tuesday, 2-2:30 pm; Wednesday 11:30-12:00 noon

Comet and Science Gateways

  • Summary: Brief talk highlighting the gateway program generally and Comet successes specifically.

  • Presenters: Nancy Wilkins-Diehr

  • Location: SDSC Booth 823

  • Time: Tuesday 10 am -12 noon. Staff will be on hand for Comet and gateway-related discussions throughout SC15

CyberGIS Gateway

  • Summary: Introduce the CyberGIS Gateway and the hybrid architecture and geospatial computational capabilities of the ROGER supercomputer

  • Presenters: Yan Liu and Shaowen Wang

  • Location: NCSA booth #1733

  • Time: Tuesday 3-4pm, demo and presentation.

Globus

  • Summary: Globus has a booth and members of the Globus team will be available to discuss and demo all things Globus: Transfer, Sharing, Publication, Auth, etc.  Also, we will be demoing some new features in development that will be coming soon.

  • Presenters: Stu Martin, Steve Tuecke, Vas Vasiliadis, Rachana Ananthakrishnan

  • Location: Globus booth (#383)

  • Time: Anytime showfloor is open.

Neuroscience Gateway

  • Summary: Brief talk about Neuroscience Gateway (NSG)

  • Presenters: Subha Sivagnanam and Amit Majumdar

  • Location: SDSC Booth

  • Time: Tuesday 10-12PM. This is a part of series of talks at SDSC booth.

Neuroscience Gateway Poster

  • Summary: Introducing Neuroscience Gateway and NEURON performance on MICs. http://sc15.supercomputing.org/schedule/event_detail?evid=post314

  • Presenters: Subha Sivagnanam, Kenneth Yoshimoto

  • Location: Level 4 lobby

  • Time: Tuesday 5-7PM.

DiaGrid: A Science Gateway for Instant Access to Computing

  • Summary: DiaGrid (www.diagrid.org) is a science gateway to help the science community compute, share tools and data, and connect to other users. It is free to sign up and access XSEDE and campus computing resources instantly through hosted tools (e.g., GROMACS, NAMD, BLAST, ParaView, RStudio).  Users can launch simulations and analyze results with an ordinary browser without downloading, compiling, or installing any software. Tool developers can follow an automated process to create, install, test, and publish tools on DiaGrid. Training accounts are available for teaching and training events. DiaGrid also hosts other domain specific gateway, including the XSEDE NLACE Gateway (for biomechanical imaging analysis) and BLAST gateway (work in progress).

  • Presenters: Chris Thompson, Carol Song, Lan Zhao

  • Location: Purdue University Booth 303

  • Time:  Demonstration Tuesday, 2:00 pm - 4:00 pm, Wednesday 2:00 pm - 4:00 pm, Thursday 1:30 - 2:30 pm.

WaterHUB: A Science Gateway for Water Research and Education

  • Summary: WaterHUB is a global communication and research platform for research, education, knowledge sharing and global outreach. Built on the HUBzero cyberinfrastructure for scientific collaboration, WaterHUB develops and makes available modeling and data sharing tools used in water related research and education activities. Among its tools, the SWATShare online application not only allows users to run and calibrate their SWAT models, it also supports sharing of models and output data visualization through web browsers. In addition, WaterHUB will also use other national level efforts such as the Consortium of Universities for Advancement of Hydrologic Science Inc's (CUAHSI) Hydrologic Information System (HIS) web services for accessing public domain and research level point observations for streamflow, precipitation and water quality.

  • Presenters: Carol Song, Lan Zhao, Larry Biehl

  • Location: Purdue University Booth 303

  • Time:  Demonstration Tuesday, 2:00 pm - 4:00 pm, Wednesday 2:00 pm - 4:00 pm, Thursday 1:30 - 2:30 pm.

Science Gateways of Notre Dame’s Center for Research Computing

  • Summary: The Center for Research Computing (CRC) at the University of Notre Dame is an innovative and multidisciplinary research environment that supports collaboration to facilitate discoveries in science and engineering, the arts, humanities and social sciences, through advanced computation, data analysis and other digital research tools. At the University of Notre Dame booth, the CRC will showcase their research efforts, collaborations with campus partners and developed science gateways utilizing computational resources to make advancements. The booth will feature multiple research projects from shock wave processing to hurricane modeling, virtual reality demos, and 3D scanning and 3D printing. CRC will also feature its involvement in the National Data Service through the series of demos of the NDS Dashboard - a tool enabling linking your data and software to publications, and publishing them all together for sharing and reuse.  

  • Presenters: Christopher Sweet, Charles Vardeman and Sandra Gesing

  • Location: Booth #393

  • Time:  Open during exhibit floor hours

WORKS 2015 Workshop (Workflows in Support of Large-Scale Science)

  • Summary: This workshop focuses on the many facets of data-intensive workflow management systems, ranging from job execution to service management and the coordination of data, service and job dependencies. The workshop therefore covers a broad range of issues in the scientific workflow lifecycle that include: data intensive workflows representation and enactment; designing workflow composition interfaces; workflow mapping techniques that may optimize the execution of the workflow; workflow enactment engines that need to deal with failures in the application and execution environment; and a number of computer science problems related to scientific workflows such as semantic technologies, compiler methods, fault detection and tolerance.

  • Presenters: see program at http://works.cs.cf.ac.uk/program.php

  • Location: Hilton 408

  • Time:  Sunday, 9:00am - 5:30pm

Big Data Analytics: Challenges and Opportunities (BDAC-15)

  • Summary: The 6th International Workshop on Big Data Analytics: Challenges, and Opportunities (BDAC-15) provides an international platform to share and discuss recent research results in adopting high-end computing including clouds and distributed computing resources for petascale - exascale data frameworks, analytics, and visualization. In the last ten years, computing capability has increased many-fold, and correspondingly data volumes have grown by an even larger amount. Many traditional application domains have now become data intensive. It is estimated that organizations with high-performance computing infrastructures and data centers are doubling the amount of data that they are archiving every year. Recent advances in computing architectures require that middleware and application software be re-engineered to fully exploit heterogeneous resources, memory hierarchies, and I/O pipelines. This workshop intends to bring together members of academia, government and industry to discuss new and emerging trends in computing architectures, programming models, I/O services, and data analytics. This workshop will also identify the greatest challenges in embracing high-end computing infrastructure for scaling I/O and algorithms to extreme scale datasets.

  • Presenters: Sandra Gesing (full program TBA)

  • Location: Hilton Salon J

  • Time:  Monday, 9:00am - 5:30pm

Characterizing Extreme-Scale Computational and Data-Intensive Workflows BOF
  • Summary: In the future, automated workflows will need to couple tasks and transfer data across millions of cores on a variety of new architectures. This BoF will explore how to characterize workflows connecting simulations, data analytics, and physical instruments. We will examine mini workflows for this exploration. Analogous to mini apps for individual codes, mini workflows allow benchmarking, performance testing, and enable the co-design of hardware and software systems for executing full-scale workflows in the future. In particular, mini workflows can be used to explore how to bridge the gap between workflows on HPC and distributed systems.
  • Presenters: Edward Seidel, Franck Cappello, Tom Peterka
  • Location: Room 13A
  • Time: Tuesday, 3:30pm - 5:00pm

Pegasus Workflow Management System Talk at the DataCloud Workshop

  • Summary: Many science applications involve large-scale, multi-stage simulation and data analysis pipelines. The amount of computation and data involved in these pipelines requires scalable workflow management systems that are able to reliably and efficiently coordinate and automate data movement and task execution on distributed computational resources such as campus clusters, national cyberinfrastructures, and commercial and academic clouds. In this talk I will describe the design and functionality of the Pegasus Workflow Management System. Pegasus has been used for more than twelve years by scientists in a wide variety of domains, including astronomy, seismology, bioinformatics, physics and others. This talk will explain how Pegasus achieves reliable, scalable workflow execution across a wide variety of computing infrastructures.

  • Presenters: Gideon Juve

  • Location: Hilton 412

  • Time:  Sunday, 11:30am - 12:00pm

Swift Parallel Scripting: Making Supercomputers and Clouds Easier to Use for Scientific Workflow

  • Summary: The Swift parallel programming language allows users to perform large-scale simultaneous runs of simulations and data analyses more efficiently and with less user effort, on cloud, cluster and supercomputer resources. Users write what looks like ordinary serial scripts; Swift automatically spreads the work expressed in those scripts over as many parallel CPUs as the user has available. Swift efficiently automates critical functions which are hard, costly, and unproductive to do manually:

    • implicit parallelization using functional dataflow
    • data transport and distribution of work across diverse systems
    • failure recover and error handling
    Swift is both portable and fast. It provides a uniform way to run scientific and engineering application workflows over diverse parallel multicore PCs and clouds. Swift is used in applications in materials, chemistry, biology, earth systems science, power grid modeling, architectural design and urban planning.

     SC15 paper presentations and demos featuring Swift:
  • WORKS paper (Wozniak et al.) - Sunday, 2:30 PM, Hilton 408
    • Interlanguage Parallel Scripting for Distributed Memory Scientific Computing
  • MTAGS paper (Ozik et al.) - Sunday, 3:00 PM, Hilton Salon G (4th floor)
    • Many Resident Task Computing in Support of Dynamic Ensemble Computations
  • ISAV paper (Dorier et al) - Monday, 3:30 PM, Hilton Salon D
    • Lessons Learned from Building In Situ Coupling Frameworks 
  • DoE Booth Demos (Wed, Thu at 11:00 AM)


Pittsburgh Supercomputing Center's Bridges 

  • Summary: Bridges will be featured at the following times and places (and maybe more).  
  • Location:  PSC booth.
  • Time:  All exhibit times at the PSC booth, featuring a Bridges prototype (“HotMetal”) and a Bridges virtual tour. Additional presentations: Monday 7:15pm, Intel booth: Intel Omni-Path Architecture announcement. Tuesday 4:00pm, Intel booth: Fabric panel. Thursday 10:30am, NVIDIA booth Thursday 11:30am, HP booth