Gateway Update for
AUS
Nancy Wilkins
-
Diehr
TeraGrid Area Director for Science
Gateways
wilkinsn@sdsc.edu
AUS telecon, January 28, 2010
How did the Gateway program develop?
A natural result of the impact of the internet on worldwide
communication and information retrieval
•
Implications on the conduct of science are still evolving
–
1980’s, Early gateways, National Center for Biotechnology Information BLAST
server, search results sent by email, still a working portal today
–
1989 World Wide Web developed at CERN
–
1992 Mosaic web browser developed
–
1995 “International Protein Data Bank Enhanced by Computer Browser”
–
2004 TeraGrid project director Rick Stevens recognized growth in scientific
portal development and proposed the Science Gateway Program
–
Today, Web 3.0 and programmatic exchange of data between web pages
•
Simultaneous explosion of digital information
–
Growing analysis needs in many, many scientific areas
–
Sensors, telescopes, satellites, digital images and video,
–
#1 machine on Top500 today over 1000x more powerful than
all combined
entries on the first list in 1993
AUS telecon, January 28, 2010
Only 18 years since the release of Mosaic!
Why are gateways worth the effort?
•
Increasing range of
expertise needed to tackle
the most challenging
scientific problems
–
How many details do you want
each individual scientist to need
to know?
•
PBS, RSL, Condor
•
Coupling multi
-
scale codes
•
Assembling data from multiple
sources
•
Collaboration frameworks
AUS telecon, January 28, 2010
#! /bin/sh
#PBS
-
q dque
#PBS
-
l nodes=1:ppn=2
#PBS
-
l walltime=00:02:00
#PBS
-
o pbs.out
#PBS
-
e pbs.err
#PBS
-
V
cd /users/wilkinsn/tutorial/exercise_3
../bin/mcell nmj_recon.main.mdl
+(
&(resourceManagerContact="tg
-
login1.sdsc.teragrid.org/jobmanager
-
pbs")
(executable="/users/birnbaum/tutorial/bin/mcell")
(arguments=nmj_recon.main.mdl)
(count=128)
(hostCount=10)
(maxtime=2)
(directory="/users/birnbaum/tutorial/exercise_3")
(stdout="/users/birnbaum/tutorial/exercise_3/globus.out")
(stderr="/users/birnbaum/tutorial/exercise_3/globus.err")
)
=======
# Full path to executable
executable=/users/wilkinsn/tutorial/bin/mcell
# Working directory, where Condor
-
G will write
# its output and error files on the local machine.
initialdir=/users/wilkinsn/tutorial/exercise_3
# To set the working directory of the remote job, we
# specify it in this globus RSL, which will be appended
# to the RSL that Condor
-
G generates
globusrsl=(directory='/users/wilkinsn/tutorial/exercise_3')
# Arguments to pass to executable.
arguments=nmj_recon.main.mdl
# Condor
-
G can stage the executable
transfer_executable=false
# Specify the globus resource to execute the job
globusscheduler=tg
-
login1.sdsc.teragrid.org/jobmanager
-
pbs
# Condor has multiple universes, but Condor
-
G always
uses globus
universe=globus
# Files to receive sdout and stderr.
output=condor.out
error=condor.err
# Specify the number of copies of the job to submit to the
condor queue.
queue 1
Gateways democratize access to high
end resources
•
Almost anyone can investigate scientific questions using high
end resources
–
Not just those in the research groups of those who request allocations
–
Gateways allow anyone with a web browser to explore
•
Opportunities can be uncovered via google
–
At 11, my son discovered nanoHUB.org while his class was studying
Bucky Balls in science class
•
Foster new ideas, cross
-
disciplinary approaches
–
Encourage students to experiment
•
Multi
-
disciplinary computational linguistics course at U Chicago uses Social
Informatics DataGrid (SIDGrid) gateway
•
But used in production too
–
Significant number of papers resulting from gateways including
GridChem, nanoHUB
–
Scientists can focus on challenging science problems rather than
challenging infrastructure problems
AUS telecon, January 28, 2010
Today, there are approximately 35
gateways using the TeraGrid
AUS telecon, January 28, 2010
Not just ease of use
What can scientists do that they
couldn’t do previously?
•
Linked Environments for Atmospheric Discovery (LEAD)
-
access to radar data
•
National Virtual Observatory (NVO)
–
access to sky surveys
•
Ocean Observing Initiative (OOI)
–
access to sensor data
•
PolarGrid
–
access to polar ice sheet data
•
SIDGrid
–
expensive datasets, analysis tools
•
GridChem
–
coupling multiscale codes
•
How would this have been done before gateways?
AUS telecon, January 28, 2010
What makes a gateway a
TeraGrid
gateway?
•
TeraGrid gateways
use
TeraGrid resources
•
Are they all developed by TeraGrid?
–
No,
we don’t make gateways the gateways you use, we make
the gateways you use better
–
The strength of the program lies in the development of end user
interfaces by those in the community
•
TeraGrid does provide staff to assist with gateway use of the
resources
–
Anyone can request support via the same peer review process used to
request CPU hours or a data allocation
–
Works just like AUS
•
Staff assigned to incoming projects
AUS telecon, January 28, 2010
Gateway
-
AUS crossover
•
Several projects requesting multiple types of support today
–
Gridchem
–
Ultrascan
•
Request for code porting/compiling, performance, parallelization, but also
database and web server hosting and improved fault tolerance for grid
software
•
Please let me know if any of the researchers you work with
have gateway needs
–
We can evaluate needs and assign staff with the right expertise to help
–
We’ll do the same if we see any requests for porting, scaling,
optimization support
AUS telecon, January 28, 2010
Some history behind gateway
allocations
•
Individual and community allocations written into policy in
2002!
–
Dick Crutcher, John Towns, Phil Andrews, Nancy author white paper
•
Today, the xRAC accepts and reviews proposals in four
general categories
–
Individual investigators
–
Large research collaborations (e.g., MILC consortium)
–
Community Projects (e.g., NEES)
–
Community Services (e.g., ROBETTA)
•
The general requirements for proposals of all four types
remain largely the same.
AUS telecon, January 28, 2010
I. Research Objectives
•
Traditional proposals
–
Describe the research activities to be pursued
–
Keep it short: You only need enough detail to support the methods and
computational plan being proposed.
•
Community proposals
–
Describe the
classes
of research activities that the proposed effort will
support.
–
Keep it short, but provide enough detail to support the rest of the
proposal
•
TIP: Reviewers don’t want to read the proposal you
submitted to NSF/NIH/etc, but they will notice whether you
have funding to pursue these activities.
AUS telecon, January 28, 2010
II. Codes and Methods
•
Very similar between traditional and community proposals.
•
More significant if using ‘home
-
grown’ codes.
–
If using widely known third
-
party codes (e.g., NAMD, CHARMM,
AMBER), you can cut some corners here, although you should explain
why you chose this code over alternatives.
•
Provide performance and scaling data on problems and test
cases similar to those you plan to pursue.
–
Or that you expect the community to pursue.
–
Describe why this code is a good fit for the resource(s) requested
and/or list acceptable alternatives.
•
Ideally, provide performance and scaling data collected by
you for the specific resource(s) you are requesting
–
This gives reviewers additional confidence that you know what you’re
doing.
AUS telecon, January 28, 2010
III. Computational Plan
•
Traditional proposals
–
Explicitly describe the problem cases you will examine
•
BAD: “…a dozen or so important proteins under various conditions…”
•
GOOD: “…7 proteins [listed here; include scientific importance of these
selections somewhere, too]. Each protein will require [X] number of runs,
varying 3 parameters [listed here] [in very specific and scientifically
meaningful ways]…”
•
Community proposals
–
Explicitly describe the typical use
-
case(s) that the gateway supports
and the type of runs that you expect individual users to make
–
Describe how you will help ensure that the community will make
scientifically meaningful runs (if applicable)
•
BAD: “…the gateway lets users run NAMD on TeraGrid resources…”
•
BETTER: “…we expect most users to run NAMD jobs on [systems like this]…”
•
BETTER STILL: “…the gateway allows users to run NAMD jobs on up to 128
processors on problem sizes limited [in some fashion]…”
AUS telecon, January 28, 2010
IV. Justification of SUs (Traditional)
•
Traditional proposals
–
If you’ve done sections II and III well, this section should be a
straightforward math problem
–
For each research problem, calculate the SUs required based on runs
defined in III and the timings in section II, broken out appropriately by
resource
•
Reasonable scaling estimates from test
-
case timing runs to full
-
scale
production runs are acceptable.
–
Clear presentation here will allow reviewers to cut time in a rational
fashion
AUS telecon, January 28, 2010
IV. Justification of SUs (Community)
•
Community proposals
–
The first big trick: Calculating SUs when you don’t know the precise
runs to be made
a priori
.
–
In Year 2 and beyond
•
Start with an estimate of total usage based on prior year’s usage patterns
and estimate for coming year’s usage patterns (justify in Section V).
•
From this information, along with data from sections II and III, you can
come up with a tabulation of SU estimates.
–
Year 1 requires bootstrapping
•
Pick conservative values (and justify them) for the size of the community
and runs to be made, and calculate SUs.
•
TIP: Start modestly. If you have ~0 users, don’t expect the xRAC to believe
that you will get thousands (or even hundreds) in the next year.
AUS telecon, January 28, 2010
V. Add’l Considerations (Community)
•
For community proposals, these components can provide
key details:
–
Community Support and Management Plan
•
Instead of staff/experience
•
You may want to include brief description of gateway interface, the fact that
it has been used for production work, relevant development effort
—
in
terms of how it helps community burn SUs.
•
If you have a plan for growing the user community, for “graduating” users
from the gateway to their own MRAC awards, it would be good to mention.
If you somehow regulate “gateway hogs,” describe that.
–
Progress report:
Provide details of the actual user community and
usage patterns seen in the prior award period.
•
List manuscripts published, accepted, submitted or in preparation, thanks to
this service. Helps convince xRAC that SUs haven’t gone down a black hole.
–
Local computing environment, Other HPC resources:
Same as
for traditional proposals.
AUS telecon, January 28, 2010
3 steps to connect a gateway to
TeraGrid
•
Request an allocation
–
Only a 1 paragraph abstract
required for up to 200k CPU hours
•
Register your gateway
–
Visibility on public TeraGrid
page
•
Request a
community
account
–
Run jobs for others via your portal
•
Staff support is available!
•
www.teragrid.org/gateways
AUS telecon, January 28, 2010
Tremendous Opportunities Using the Largest Shared Resources
-
Challenges too!
•
What’s different when the resource doesn’t belong just to
me?
–
Peer
-
reviewed requests for resources
–
Resource discovery, fault tolerance
–
Accounting
•
Must keep track of who’s used what on the gateway
•
Attribute
-
based authentication for TeraGrid jobs
–
Security
–
Software registry for TeraGrid
•
Tremendous benefits at the high end, but even more work
for the developers
•
Potential impact on science is huge
–
Small number of developers can impact thousands of scientists
–
But need a way to train and fund those developers
AUS telecon, January 28, 2010
What are we working on now?
•
Arroyo
–
Adaptive optics corrections for
telescopes
•
GridChem
•
Ultrascan
–
Analysis of ultracentrifugation
experiments
•
Earth System Grid
-
Community
Climate System Model
•
GISolve
•
SimpleGrid
•
Social Informatics DataGrid
•
Open Life Sciences Gateway
•
Pyrosequencing
•
RENCI Bioportal
•
Asteroseismology Modeling Portal
•
Uintah
•
Gateway software
listing
•
Investigate use of TG
for overflow OSG jobs
–
RENCI bioportal,
nanoHUB, using both
resources
•
Common treatment of
community accounts
•
Attribute
-
based
authentication
AUS telecon, January 28, 2010
From the Condor flock to Kraken
From simple interfaces to complex video analysis
Diverse goals keep program interesting
•
AMP gateway, http://amp.ucar.edu
–
Derive the properties of Sun
-
like stars from observations of their
pulsation frequencies
–
Kepler mission will use asteroseismology to determine precise absolute
sizes of the potentially habitable Earth
-
like planets
–
Simple interface, few input parameters, very large CPU consumption
on Kraken, database kept so simulations aren’t re
-
run
•
Robetta gateway, http://www.robetta.org
–
Protein structure prediction
–
Provides access to Dr. David Baker’s award
-
winning Rosetta code
–
2M hours on Purdue’s Condor pool, very successfully using this time,
reduced calculation backlog to zero recently running 300 jobs
simultaneously
•
Social Informatics DataGrid, https://sidgrid.ci.uchicago.edu
–
Slide later
AUS telecon, January 28, 2010
Gateway activities in PY6 aka the
extension
•
Helpdesk support expanded
–
From .2 FTE in PY5 to 1.7 in Extension [NCSA, Purdue]
•
Helpdesk and Condor support, new GIS communities, SimpleGrid extensions
•
Accounting
–
Improved views for gateways now that we have attributes [TACC]
•
Community accounts
–
Continued work toward improved standardization [NICS]
•
Prebuilt VMs with gateway software
–
OGCE, SimpleGrid [IU, NCSA]
•
Online tutorials with CI Tutor and the EOT team
–
OGCE, SimpleGrid [IU, NCSA]
•
More example
-
based documentation
–
Less talk, more action, short videos, based on user feedback [NCSA,
SDSC]
AUS telecon, January 28, 2010
What else is exciting?
•
SGW funding a Cyber
-
GIS workshop in conjunction with the
UCGIS meeting in DC in February
–
Co
-
led by Shaowen Wang at NCSA and Nancy
–
Approved by UCGIS board after a lengthy voting process
–
Winter UCGIS meeting will focus on CI and includes a White House
briefing
•
Workshop attendees welcome to attend the briefing
•
Briefing will include very short recap of the workshop
–
Expected outcome of the workshop
•
Increased awareness of CI resources for GIS researchers
•
Increased visibility for TeraGrid
•
New partnerships for TeraGrid, UCGIS, and other pertinent organizations
•
Workshop report
•
Interesting collaborative proposal ideas
•
Potential future publications
AUS telecon, January 28, 2010
A few gateways in detail
AUS telecon, January 28, 2010
SCEC Gateway used to produce realistic
hazard map
•
Probabilistic Seismic Hazard
Analysis (PSHA) map for California
–
Created from Earthquake Rupture
Forecasts (ERC)
•
~7000 ruptures can have 415,000
variations
•
Warm colors indicate regions with
a high probability of experiencing
strong ground motion in the next
50 years
•
Ground motion calculated using
full 3
-
D waveform modeling for
improved accuracy
–
Results in significant CPU use
AUS telecon, January 28, 2010
SCEC: Why a gateway?
•
Calculations need to be done for each of the hundreds of
thousands of rupture variations
–
SCEC has developed the “CyberShake computational platform”
•
Hardware, software and people which combine to produce a useful scientific
result
–
For each site of interest
-
two large
-
scale MPI calculations and hundreds
of thousands of independent post
-
processing jobs with significant data
generation
»
Jobs aggregated to appear as a single job to the TeraGrid
»
Workflow throughput optimizations and use of SCEC’s gateway
“platform” reduced time to solution by a factor of three
–
Computationally
-
intensive tasks, plus the need for reduced time to
solution is a priority make TeraGrid a good fit
AUS telecon, January 28, 2010
Source: S. Callahan et.al. “Reducing Time
-
to
-
Solution Using Distributed High
-
Throughput Mega
-
Workflows
–
Experiences from SCEC CyberShake”.
GridChem
•
Understanding molecular structure and function increasingly important in
many fields
–
Materials for electronics, biotechnology, medical devices, pharmaceutical
design
•
GridChem provides reliable infrastructure for computational chemists
–
NSF Middleware Initiative (NMI) project
•
Requested and received advanced support from TeraGrid
–
Addressing issues which benefit all gateways, support team led by IU
•
Common user environments for domain software access
•
Standardized licensing
•
Application performance characteristics
•
Incorporation of additional data handling tools and data resources
•
Fault tolerant workflows
•
Scheduling policies for community users
•
Remote visualization
AUS telecon, January 28, 2010
GridChem: Why a gateway?
•
Integrates high end resources in a
desktop environment
–
Client
-
server approach allows work
to continue while disconnected
(plane flights)
–
Ability to monitor jobs across sites
–
Access to individual allocations
–
In the future, linkage of multi
-
scale
packages
•
Focus on chemistry research
without learning the intricacies of
each system
–
Time limits, nodes, processors,
memory, disk space, etc
AUS telecon, January 28, 2010
Robetta Gateway
Protein structure prediction with an award
-
winning code
•
Protein structure prediction is among many important
problems in bioinformatics.
•
The Rosetta code, from the David Baker laboratory, has
performed very well at CASP (Critical Assessment of
Techniques for Protein Structure Prediction) competitions
–
Available for use by any academic scientist via the Robetta server
•
Robetta developers able to use TeraGrid’s existing gateway
infrastructure, including community accounts and Globus
–
This very successful group needed no additional TeraGrid assistance to
incorporate TeraGrid resources into the Robetta gateway
•
Google scholar reports 601 references to the Robetta
gateway, including many PubMed publications
AUS telecon, January 28, 2010
Robetta: Why a gateway?
•
Bioinformatics has long
history of web
-
based
services
–
NCBI Blast server from the
1990s
•
Easy input from the
web
•
Access to top modeling
code for all researchers
AUS telecon, January 28, 2010
Social Informatics Data Grid
Collaborative access to large, complex datasets
•
SIDGrid is unique among
social science data archive
projects
–
Focused on streaming data
which change over time
•
Voice, video, images (e.g. fMRI),
text, numerical (e.g. heartrate,
eye movement)
–
Provides the ability to
investigate multiple datasets,
collected at different time
scales, simultaneously
•
Large datasets result
–
Sophisticated analysis tools
AUS telecon, January 28, 2010
http://www.ci.uchicago.edu/research/files/sidgrid.mov
SIDGrid: Why a gateway?
•
Social scientists have traditionally
worked in isolated labs without the
capability to share data or insights
with others.
–
Data that is expensive to collect can
now be shared with others
–
Geographically distant researchers can
collaborate
–
Complex analysis tools and workflows
available for all
–
Researchers have access to high
performance computational resources
•
TeraGrid used for
computationally
-
intensive tasks
such as media transcoding
algorithms for pitch analysis of
audio tracks and fMRI image
analysis
AUS telecon, January 28, 2010
Source: Dr. Steven Boker, Notre Dame
Viewing multimodal data like a
symphony conductor
•
“Music
-
score” display and
synchronized playback of video
and audio files
–
Pitch tracks
–
Text
–
Head nods, pause, gesture
references
•
Central archive of multi
-
modal
data, annotations, and analyses
–
Distributed annotation efforts by
multiple researchers working on a
common data set
•
History of updates
•
Computational tools
–
Distributed acoustic analysis using
Praat
–
Statistical analysis using R
–
Matrix computations using Matlab
and Octave
AUS telecon, January 28, 2010
Source: Studying Discourse and Dialog with SIDGrid, Levow, 2008
Uintah
•
Product of a Utah's DOE ASC Center
–
C
-
SAFE
•
Component based framework for solving PDEs on structured
AMR grids.
•
Computations are expressed as tasks based on inputs and
outputs for each patch in the structured grid.
•
Tasks are organized in a task graph and assigned processing
resources by the built in scheduler.
•
Load balancing is achieved by a fast space filling curve
algorithm for the patches.
•
Primary components are CFD (Arches and ICE), Solid
Mechanics (MPM) and Fluid
-
Structures (MPM
-
ICE)
•
www.uintah.utah.edu
AUS telecon, January 28, 2010
Source: John Schmidt, U Utah
Uintah CFD Components
Industrial Flare
Simulation using Arches
Component
Prediction of flame shape
and tilt using LES
Prediction of pollutant
emissions
AUS telecon, January 28, 2010
Source: John Schmidt, U Utah
Fluid Structure Interaction
Microscale Fluid
Structure Interaction
using MPM
-
ICE
Component
Array of pins undergoing
deformation which
influences heat transfer
Potential application in
CPU cooling
AUS telecon, January 28, 2010
Source: John Schmidt, U Utah
Solid Mechanics Simulation
Shape charge detonating
forming a jet of particles
which penetrate a steel
target using the MPM
solid mechanics
component.
AUS telecon, January 28, 2010
Source: John Schmidt, U Utah
Uintah Science Gateway
•
Manage the end to end job submission and data
management for TeraGrid resources.
•
Use the Django framework and Postgresql for the front
-
end
calling Globus scripts to interact with the back
-
end TeraGrid
machines. Strictly web based.
•
Target new and existing users.
•
New users will take advantage of web front end for quickly
getting up to speed on TeraGrid resources.
•
Provide existing users with various data management and
work flow features.
AUS telecon, January 28, 2010
Source: John Schmidt, U Utah
Scaling on Kraken
Uintah scales to very
large processor core
counts
Scaling for an AMR MPM
-
ICE (Fluid Structure
Interaction) problem
demonstrating both fixed
and increasing problem
size
AUS telecon, January 28, 2010
Source: John Schmidt, U Utah
Scaling on Kraken
An AMR CFD example for
the ICE component for
large processor core
counts
Scales fairly well on
Kraken both for fixed and
increasing problem size
AUS telecon, January 28, 2010
Source: John Schmidt, U Utah
Future Technical Areas
•
Web technologies change fast
–
Must be able to adapt quickly
•
Gateways and gadgets
–
Gateway components incorporated
into any social networking page
–
75% of 18 to 24 year
-
olds have
social networking websites
•
iPhone apps?
•
Web 3.0
–
Beyond social networking and
sharing content
–
Standards and querying interfaces
to programmatically share data
across sites
•
Resource Description Framework (RDF),
SPARQL
AUS telecon, January 28, 2010
Gateways can further investments in
other projects
•
Increase access
–
To instruments, expensive data collections
•
Increase capabilities
–
To analyze data
•
Improve workforce development
–
Can prepare students to function in today’s cross
-
disciplinary world
•
Increase outreach
•
Increase public awareness
–
Public sees value in investments in large facilities
–
Pew 2006 study indicates that half of all internet users have been to a
site specializing in science
–
Those who seek out science information on the internet are more likely
to believe that scientific pursuits have a positive impact on society
AUS telecon, January 28, 2010
Sustainability is the key though
•
Scientists will tie their research to a tool that they aren’t
convinced has a long life
–
But, all projects can’t be funded for the long term
•
Nancy currently leading small 2
-
year EAGER study to look at
the characteristics of gateways that warrant sustained
funding
–
Working with Katherine Lawrence, U Michigan School of Information
AUS telecon, January 28, 2010
Tremendous Potential for Gateways
•
In only 18 years, the Web has fundamentally changed
human communication
•
Science Gateways can leverage this amazingly powerful tool
to:
–
Transform the way scientists collaborate
–
Streamline conduct of science
–
Influence the public’s perception of science
•
Reliability, trust, continuity are fundamental to truly change
the conduct of science through the use of gateways
–
High end resources can have a profound impact
•
The future is very exciting!
AUS telecon, January 28, 2010
AUS telecon, January 28, 2010
Please let us know if you see gateway interest
from researchers you work with.
Thanks for the opportunity to present.
Nancy Wilkins
-
Diehr,
wilkinsn@sdsc.edu
www.teragrid.org
Enter the password to open this PDF file:
File name:
-
File size:
-
Title:
-
Author:
-
Subject:
-
Keywords:
-
Creation Date:
-
Modification Date:
-
Creator:
-
PDF Producer:
-
PDF Version:
-
Page Count:
-
Preparing document for printing…
0%
Comments 0
Log in to post a comment