1. Integrating Scientific Tools and Web Portals
Kevin Chen

2. Overview What is DiaGrid? Supporting Science Tools Gromacsimum
Interactive Parallelization Tool (IPT)

3. What Is DiaGrid ? R Java Data Science C++ To the users DiaGrid is…
A HUBzero portal for collaboration and community building.
Scientific Software-as-a-Service with easy access to a vast set of computing resources.
A remotely accessible home for research.
To the developers DiaGrid is…
A federation of a vast set of computing resources.
A pipeline for the whole development process.
Managed deployment straight to users.
A support platform for communicating directly with end users.
Python
Java
Data
Results R
Science
C++

4. The Hardware
Large high-throughput and distributed network, available through HT Condor.
Utilizes spare cycles from:
Campus lab workstations
Departmental desktop computers
XSEDE Resources
Community clusters at Purdue
Coates, Rossmann, Hansen, & Carter
Nearly 100 million jobs run to date!

5. Building Communities
Users have access to a full inbox-style messaging system.

6. Supporting Science
The true power of the HUBzero platform lies in transparently bridging the gap between the grid and the web.

7. Supporting Science DiaGrid applications execute within a VM.
Imbedded Java applet/Html5 interfaces give users access from any web browser.
No local installation of anything is required at user’s location.

8. Supporting Science
Create
Tool
From creation to release, tool development has a full lifecycle managed by the DiaGrid site.
Published version continues to live on site for users while next version is still in testing.
Upload/Edit
Sources
Stage & Test
Approve?
Viewable
to Devs
Publicly
Available
Published!

9. Supporting Science
DiaGrid.org
The HUBzero team has created the “submit” shell command to abstract grid access for tool developers.
Tools run a subprocess through “submit” to handle all their grid computation needs.
Utilizes Pegasus engine for HT Condor on resources.
Tool Session
Tool
Submit
Pegasus
HT Condor
CPU
CPU
CPU
CPU
CPU

10. What is GROMACSIMUM ?
GROMACSIMUM is a graphical user interface for popular molecular dynamics package, GROMACS. GROMACSIMUM’s combined accuracy and speed make possible long time molecular dynamics simulations.
Seamlessly integrated with newly developed GUI interfaces, GROMACSIMUM provides comprehensive setup, simulation, analysis and job submission tools.
With its highly visual, easy-to-use interface, GROMACSIMUM guides you through each stage of the MD simulation process; in guiding your project workflow, its user friendly environment directs you to a successful outcome.
Users inexperienced in MD can work along prepared workflows, while experts may enjoy a significant relief from the tedium of typing and scripting.  

11. Easy-to-use, instantly interactive
Newly incorporated workflow management system.
A significant relief from tedium of typing and scripting
Easy job submission mechanism.
Dedicated HPC resources.

12. Easy-to-use, instantly interactive

13. Easy-to-use, instantly interactive

14. Data analysis and decision making
Plotting, editing, visual data mining and filtering to identify key interactions.
Trajectory viewing and analysis.
Free energy surfaces, Principal component, Clustering and Secondary structure analysis,.
Seamless support for Xmgrace, VMD, PyMOL and other 3rd party software packages.

15. Usage: Overview

16. Interactive Parallelization Tool (IPT)16

17. Interactive Parallelization Tool
Interactive Parallelization Tool (IPT) is a semi-automatic tool that can be used by domain experts and students for transforming certain classes of existing applications into multiple parallel variants.
The parallel programming paradigms that are currently supported by IPT are MPI,OpenMP, and CUDA.
The supported base languages are C and C++. 

18. Interactive Parallelization Tool
GUI to IPT Workspace on DiaGrid Hub showing the parallelization of the program named “circuit_serial.cpp”

19. Interactive Parallelization Tool

20. Looking to the Future…
Any other future projects?

21. Questions? 21