NSF Cyberinfrastructure User Survey 2005

NSF Cyberinfrastructure User Survey 2005

To ensure the effective support of its broad user community, NSF-supported centers have conducted annual user surveys for many years. In August and September 2005, the annual survey was conducted for the first time as a joint activity of the Cyberinfrastructure Partnership (SDSC and NCSA) and the TeraGrid. This year's survey not only showed the continued success of the NSF-supported cyberinfrastructure environment, but also demonstrated the user community's general satisfaction with the resources, software and services from all NSF cyberinfrastructure sites.

The survey responses are summarized in Table 1 below. Satisfaction scores are given on a scale of 1–5, with 5 being "extremely satisfied." The table shows the top five responses for questions; unless otherwise indicated, the table shows the number of respondents and the percentage of respondents selecting that answer.

The survey was open for four weeks during August and September 2005. The 39 questions covered the areas of computational environment, grid environment and services, programming environment, software and tools, data and I/O, visualization, allocations, consulting and help desk support, training and documentation, and some optional demographic information.

We advertised the survey by sending e-mail messages to active users at NCSA, SDSC, PSC and TeraGrid. A total of 412 complete and partial responses were recorded. CIP and TeraGrid sites will make decisions about resources, services, and software based on this survey.

In the survey, users also had the opportunity to report on the scientific successes they have achieved thanks to NSF-supported resources, and about half took that opportunity. While these results are more qualitative, they do underscore the broad impact that NSF Cyberinfrastructure has across many scientific disciplines. The following user quotes are representative of the successes reported:

The survey indicates that the NSF Cyberinfrastructure programs are highly successful in meeting the overall needs of a very diverse user community. Highlights from the survey include:

The following results indicate a need to increase awareness of the opportunities and resources available through CIP and TeraGrid. The CIP and TeraGrid teams will use this input to help prioritize efforts in the coming year:

Table 1. Summary of User Survey Results

COMPUTE RESOURCES Responses Avg. Rating
ANL TeraGrid IA64 Cluster 32 3.7
Caltech TeraGrid IA64 Cluster 35 3.5
Indiana AVIDD-I64 Cluster 17 3.4
NCSA Condor Pool (radium) 15 3.7
NCSA IBM p690 (copper) 106 4.0
NCSA SGI Altix (cobalt) 64 3.9
NCSA TeraGrid IA64 Cluster 95 4.2
NCSA Xeon Cluster (tungsten) 100 4.1
ORNL Xeon Cluster 11 3.4
PSC HP Marvel (Rachel) 50 3.9
PSC TCS1 (Lemieux) 98 4.0
Purdue IBM SP 14 3.5
Purdue Linux Clusters 14 3.5
SDSC DataStar p655 67 4.2
SDSC DataStar p690 64 4.1
SDSC TeraGrid IA64 Cluster 70 4.0
TACC Cray-Dell Clstr (Lonestar) 20 3.8
TACC Sun Fire E25K (Maverick) 10 3.4
SITE, OVERALL Responses Avg. Rating
TeraGrid (overall) 148 4.0
NCSA 210 4.3
PSC 116 4.1
SDSC 131 4.2
ANL/UC 33 3.8
Caltech 29 3.4
Indiana U 19 3.4
ORNL 18 4.1
Purdue U 15 3.5
TACC 18 4.2
MASS STORAGE SYSTEMS Responses Avg. Rating
NCSA UniTree 132 4.1
PSC FAR 70 4.0
SDSC HPSS 80 4.1
TACC DMF 10 3.6
SRB 32 3.8
GridFTP-UniTree 33 3.7
GRID ENVIRONMENT AND SERVICES
Grid tools using/interested in: Experience with/plan to use grid tools:
Don't know (101; 33%)
MPICH/MPICH-G2 (97; 32%)
GridFTP (67; 22%)
None (52; 17%)
Globus command line (43; 14%)
May use w/in 2 years (88; 28%)
Starting to experiment (84; 26%)
No plan to use (80; 26%)
Experienced (57; 18%)
Advanced Portal Features desired
 (No. Responses, Interest 1-3 scale):
Metaqueue (301; 2.6)
Reserve resources (294; 2.4)
Manage staging of data (296; 2.2)
Discussions w/users (297; 2.1)
PROGRAMMING SOFTWARE AND TOOLS
Parallelization Method: Math Software/Libraries
MPI (176; 61%)
None (46; 16%)
OpenMP/MPI (39; 13%)
OpenMP (30; 10%)
Don't know (24; 8%)
None (98; 34%)
FFTW (67; 23%)
LAPACK (66; 23%)
Matlab (53; 18%)
IBM ESSL (40; 14%)
Parallel/Perform. Tools Third-party Software
None (127; 47%)
MPI_wtime() (47; 17%)
TotalView (47; 17%)
Don't know (44; 16%)
gdbx (27; 10%)
Gaussian (30)
CHARMM (12)
NAMD (12)
AMBER (10)
IDL (10)
DATA AND I/O
Software, Tools Used: Useful Persistent Disk
scp (157; 52%)
Matlab (74; 25%)
Arch stor cmds (72; 24%)
None (46; 15%)
Mathematica (40; 13%)
1 TB or less (153; 52%)
1 TB – 10 TB (103; 35%)
10 TB – 50 TB (28; 10%)
50 TB or more (12; 4%)
Persistent Disk Access by Use Data Collections
Network file sys. (193; 69%)
GridFTP (53; 19%)
Other (21; 7%)
SRB (14; 5%)
No (237; 82%)
Yes (51; 18%)
VISUALIZATION
Software Used:
None (132; 47%)
Matlab (52; 19%)
Gnuplot (50; 18%)
IDL (34; 12%)
VMD (30; 11%)
ALLOCATIONS Responses Avg. Rating
Overall Process 266 4.1
Allocations Staff 247 4.3
Online help/documentation 254 3.9
Proposal format/length 219 3.9
POPS 187 4.1
Time to get award 210 3.9
Reviewer Comments 171 3.7
CONSULTING AND HELP DESK SUPPORT
Phone Consulting Responses Avg. Rating
TeraGrid Help Desk 83 4.2
NCSA Consulting 96 4.4
NCSA Help Desk 99 4.4
SDSC Consulting 78 4.3
E-mail Consulting Responses Avg. Rating
help@teragrid.org 95 4.2
consult@ncsa.uiuc.edu 116 4.4
help@ncsa.uiuc.edu 118 4.4
consult@sdsc.edu 91 4.3
Preferred Contact Method
E-mail (217; 72%)
Telephone (44; 15%)
Interactive text chat (15; 5%)
Discussions/'bugzilla' (9; 3%)
Web forms (8; 3%)
TRAINING AND DOCUMENTATION
TRAINING Responses Avg. Rating
On-site Courses 66 4.1
Access Grid Courses 33 3.9
Video Archive 20 3.5
Web-based Courses 39 4.1
DOCUMENTATION Responses Avg. Rating
Web sites 261 3.9
User Guides 236 3.9
User News 189 3.7