NERSCPowering Scientific Discovery Since 1974

1998 User Survey Results

Respondent Summary

NERSC has completed its first user survey since its move to Lawrence Berkeley National Laboratory. The survey is intended to provide user feedback about every aspect of NERSC's operation, help us judge the quality of our services, give DOE information on how NERSC is doing, point us to areas we can improve, and show how we compare to similar facilities. We intend to repeat the survey yearly to monitor our progress.

Users rated NERSC in the areas of computational and file storage resources, account and allocations support, consulting services, documentation, and training. An area that wasn't included explicitly in the survey, but that users commented on, was software support.

There were 138 respondents. About 56 percent said they use the C90; another 54 percent use the T3E and 46 percent the J90s. There was less representation for the archival systems: 32 percent said they use CFS and 26 percent HPSS.

The three areas rated most important are network access, the center overall, and consulting services. The three areas that received the highest ratings are consulting services, network access, and account support services.

Overall our users say we provide a high level of service. On a scale of 1 (very unsatisfied) to 7 (very satisfied) our scores ranged from 5.8 for consulting services, C90 uptime, and CFS reliability to 4.4 for C90 and T3E turnaround time. Many users pointed out that NERSC is a good high-end production center, with a good user focus. Below are a sample of the comments that reflect this:

  • "Gives us access to high-end, high-capacity and capability machines without too many limitations on the way codes can be run. There is a good mix of machines, serial and parallel, and a good mix of running modes, interactive and batch."

     

  • "Supplies reliable cycles to a broad user base. Excellent user support, probably the best in the world for high-end computing."

     

  • "As a service center NERSC does a great job keeping the T3E's up and providing easily accessible information relating to hardware and software. Moreover, the intellectual quality of NERSC support and research staff is excellent and they are easy to communicate with."

     

  • "Over the years NERSC has been the most reliable source of production computing for scientific research and development, by concentrating on scientific computing ahead of computer science."

When asked what they wanted more of, most users said "more cycles". Specific software requests were also popular, as were longer T3E queues and additional batch queues in general. Areas that need improvement include:

  • turnaround time: we are addressing the need for more cycles with the NERSC-3 procurement process underway.

     

  • queue structures: a new T3E queue structure with more 4-hour queues and a new 12-hour queue will be implemented in December. The J90s were unified under NQE this October; further enhancements will be considered in early 1999.

     

  • software, visualization, and What's New web pages: these are being worked on and a new organization of NERSC information will be ready in December.

     

  • more direct communications with users: as a result of this survey we have been using email more often to communicate system changes to users.

     

  • the training process: we are now conducting teleconference classes in order to reach all users more effectively.

Below we present the results of the survey. (For the survey itself, click here.) The responses to some of the questions have been combined. To read the replies, click on the questions.

We would like to thank all the users who participated. If you wish to make any comments on these survey results, please send them to webmaster@nersc.gov.


    

Q3.   What NERSC resources do you use?

ResourceN
NERSC consulting services 86  
NERSC website 86  
C90 77  
T3E 75  
J90 64  
NERSC account support services 49  
CFS 44  
HPSS 36  
NERSC Computer Operations and Network Support 10  
Visualization server 5  
Math server 4  

    


Q4.   Your satisfaction with NERSC: ( Scale: 1 (low) to 7 (high) )

TopicNAvg.
Consulting services 126 5.87  
Network access to NERSC 118 5.70  
Account support services 124 5.67  
Overall satisfaction with NERSC 135 5.43  
C90 Overall 82 5.29  
Technical information on website 128 5.27  
T3E Overall 91 5.20  
CFS Overall 61 5.18  
HPSS Overall 64 5.09  
J90s Overall 79 4.92  
Training 103 4.91  
Allocations process 112 4.60  

How important is this to you? ( Scale: 1 (low) to 3 (high) )

TopicNAvg.
Network access to NERSC 116 2.84  
Overall satisfaction with NERSC 129 2.77  
Consulting services 128 2.70  
Technical information on website 127 2.51  
T3E Overall 99 2.45  
Account support services 123 2.31  
Allocations process 118 2.31  
C90 Overall 101 2.27  
J90s Overall 94 2.19  
CFS Overall 80 2.09  
HPSS Overall 82 2.04  
Training 117 1.99  

    


Q5.   Comments on the ratings in Q3 and Q4?

 

Comments on Section 1 Ratings (all responses here)


T3E Comments: 11 responses

  • 4 - Happy and satisfied with the T3E; positive experience
  • 2 - unsatisfied with queue policy
    • queuing system extremely awkward; doesn't support very large simulations,
    • need longer runs
  • 2 - slow turnaround (1 said especially for code development)
  • 1 - Unhappy with technical aspects of machine
    • problem with temporary memory allocations
  • 1 - unsatisfied with various aspects of programming environment
    • Fortran compiler has no VMS extensions,
  • 1 - great response time
  • 1 - quotas are too inflexible
  • 1 - really need ssh
  • 1 - need more time

 

Consulting services: 11 responses

  • 10 - very good; friendly and patient; helpful; good teachers; responsive prompt
  • 1 - uneven, depending on who your consultant is

 

General Comments: 10 responses

  • 8 - can't rate what don't use (said by some who only use a subset of services)
  • 1 - in general very pleased to have access to NERSC
  • 1 - make use of collaboratory tools to help interaction between users and users and staff

 

PVP Comments: 8 responses

  • 2 - C90 too busy (1 reasonable turnaround only in high priority queues; don't have enough allocation for that; queues should be restructured so that this isn't as big a factor)
  • 2 - happy with the C90 (1 said didn't use J90 because too slow for small jobs)
  • 1 - slow turnaround, especially for 512 MB queue which can't be done locally
  • 1 - C90 response time slow
  • 1 - poor J90 performance (worse than local workstation); will miss C90; hope J90++ a lot better
  • 1 - systems change constantly; our programs stop running; please make changes more transparent
  • 1 - don't understand J90 directory names

 

Account support services: 6 responses

  • 2 - account setup too slow (1 specified for HPSS)
  • 2 - unhappy with various parts of CUB
  • when one types setcub view, one should see the percentage of time left in the repo
  • allocations were assigned monthly, this was not convenient
  • 1 - account creation and password change process too cumbersome since all requests have to go through PI or account manager, which can create delays
  • 1 - staff are helpful

 

Allocations Process: 6 responses

  • 2 - wants allocation process to be more flexible; more than once a year; easier to ask for more time throughout year
  • 2 - wants to know evaluation criteria; allocation process seems to be a black box; requires "old boy" network support
  • 1 - T3E allocation too small
  • 1 - ERCAP interface needs improvement (too complicated)
  • 1 - process cumbersome since all requests have to go through PI

 

Technical Information on Website: 6 responses

  • 2 - sometimes hard to read/understand; needs to be better organized
  • 1 - extremely helpful
  • 1 - needs to cover more info not available through web pages
  • 1 - better batch info: how to submit runs, max memory sizes for different
  • 1 - give new users batch templates to aid the tutorial process queues, sample commands and scripts

 

File Systems: 4 responses

  • 1 - happy with CFS
  • 1 - CFS user interface is brain-dead
  • 1 - don't have enough real info on HPSS; too big an effort to get a password there
  • 1 - "Disk space management is a fiasco."

 

Network Access to NERSC: 1 response

  • 1 - network access has improved, but still too slow

 


Q6.   What does NERSC do well?

 


  1. Consulting and support staff, user focus: 47 responses
  2. Provides good access to high performance computing, good hardware or software support, well run center: 32 responses
  3. Documentation, web, online tutorials, training: 14 responses
  4. T3E support (maintenance, management): 7 responses
  5. File systems and mass storage: 6 responses
    • 3 - managing disk space (1 mentioned large homes, 1 migration)
    • 1 - HPSS is great at managing large files (100 GB to several TB!)
    • 1 - CFS access speed has improved
    • 1 - storage in general
    • 1 - backup by cfs is great
  6. C90 is a great machine: 2 responses
  7. Allocations: 2 responses
  8. Research / Scientific Computing support: 2 responses
  9. Networks: 1 response

 


 

    Summary

Q7.   In what areas should NERSC improve?

Q8.   What additional resources would you like NERSC to provide?

Q10. If you were in charge of NERSC what changes would you implement?

We combined the answers the above questions because there was a great deal of duplication in the answers. NERSC responses to these comments are included in bold.

 

 

  1. Provide more cycles / more machines: 26 responses
  2. Better software support: 20 responses
  3. Better / different batch management: 19 responses
  4. Better/different allocations process: 11 responses
  5. Improve documentation/training: 10 responses
  6. More storage, better storage interfaces, better file management, new file services: 9 responses
  7. Better accounting/charging/account management procedures: 8 responses
  8. Keep users better informed; better interactions with users: 6 responses
  9. Improve the T3E / other T3E issues: 6 responses
  10. Faster networks / better response time: 5 responses
  11. Less down time: 4 responses
  12. Provide a workstation farm; better PDSF support: 4 responses
  13. Visualization support/software: 4 responses
  14. Increase research / scientific computing support: 3 responses
  15. No changes needed: 3 responses
  16. Miscellaneous: 3 responses

 


 

     

  1. Provide more cycles / more machines: 26 responses
    • 10 - more vector cycles; keep the C90
      • "Enabling truly large-scale computing. A vector machine with tons of memory devoted entirely to HUGE jobs!"
      • "Massively parallel vector machine. At least 1024 processors."
    • 8 - more T3E or MPP cycles
      • "Our research depends very much on HPCC technology. More computing resources available to us will allow us to complete century long simulations and comparisons."
    • 7 - more cycles in general
    • 1 - SMP
      • "More emphasis on SMP architectures. I fear the MPP approach may be dead in a few years and is not relevant to many computing projects in DOE."

    NERSC is addressing this concern through the NERSC-3 procurement process.   

  2. Better software support: 20 responses
    • wanted for the T3E:
      • multithreaded Perl
      • full ports of CERN libraries
      • GNU tcsh
      • full ssh to/from with authentication
      • Concurrent Versions System (CVS)
      • more end-user oriented tools like PSPASES, pPCx, optimization tools
      • latest version of emacs
      • improve ease of use
    • wanted for the PVPs:
      • 3 requests for keeping cf77
      • f77 to f90 translator like VAST
      • faster debugger than totalview
      • MOLPRO for chemistry
      • CERNLIB
      • tcsh
      • ezget libraries from LLNL
    • wanted in general:
      • increase variety of supported software
      • better support for fixing Cray Fortran bugs
      • better file manager and editor
      • ghostview
      • pop mail server on SAS, email on SAS
      • a set of Unix scripts on line for making batch file submission easier and more user friendly
      • in-house software mode as needed to improve use of systems

    Some of the software mentioned above is already on the machines, for example: tcsh, CVS, emacs. We can't provide some of the software listed because it is no longer supported, for example, f77. We evaluate all software requests, and consider the anticipated usage and the cost.    

  3. Better / different batch management: 19 responses
    • 8 - different T3E queue management (of which 5 for longer queues and 2 for faster debug turnaround time)
      • "I am wondering whether the T3E scheduler could predict when a given job in the queue would be expected to start execution (this prediction could perhaps be included in the qstat listing); this prediction of course would be subject to change, but might be helpful for planning."
      • "you have to figure out some way that it is possible to do medium sized development work in a acceptable time frame. some bugs show up only if the job is a little larger - and in the present set-up with about 2 jobs / day ... it takes for ever to find anything."
    • 7 - better batch software
    • 4 - different PVP queue management
      • "give more priority to large job which can not be done locally"

    NERSC has developed a new proposal for the T3E queue structure which should be implemented by December 1998:

    queue name current limit new limit change
    ----------------------------------------------------------------------
    serial 4 hours 4 hours create separate pipe
    debug 0.5 hour 0.5 hour create separate pipe
    pe32 4 hours 4 hours no change
    pe64 4 hours 4 hours no change
    pe128 2 hours 4 hours 2X change
    pe256 1 hours 4 hours 4X change
    pe512 1 hours 4 hours 4X change
    gc128 12 hours 12 hours no change
    gc256 12 hours 12 hours no change
    long_128 - 12 hours new queue
    The J90 Cluster queue structure will be evaluated next. We will try to make the NQS software easier to use and understand by writing better web documentation.    

     

  4. Better/different allocations process: 11 responses
    • make it easy to get increases during the year
    • better notification to users (not just PIs)
    • make allocations more easily available to local users
    • simplify / rationalize process
    • need larger T3E allocation
    • change and make more flexible
    • ERCAP should be replaced - need to be able to use powerful editor like LateX
    • earlier deadline this year was a major inconvenience
    • make allocations based on science, not whether DOE funded
  5. Improve documentation/training: 10 responses
    • make info easier to find
    • improve writing style
    • more info
    • better info on using math and graphics libraries
    • better batch documentation
    • need Unix tutorial
    • want manual and tutorials in hand
    • more high-end training / computational science support
    • more remote site training
    • better communication of machine status
    • better communication of available software

    We do continuous work on web pages; the survey tells us we need to focus especially on the software pages and batch documentation (also on the visualization pages, but these affect fewer users). We will provide a new search engine by the end of January, 1999. We are now offering teleconference classes: see NERSC Training Classes.   

  6. More storage, better storage interfaces, better file management, new file services: 9 responses
    • 3 - better disk space management (guarantee a few megabytes per user permanently on-line; migration system unreliable; home directory policy keeps changing; more space in the work disk)
    • 2 - better user interfaces (improve or replace HPSS; put more effort into system software development; CFS is brain-dead)
    • 2 - more shared file systems (access to mcurie from newton; access to vis server)
    • 1 - more capacity
    • 1 - tape services ("we have many GBs on DLT tapes")

    HSI is being introduced as an HPSS user interface. We will continue to get more storage capacity all the time.    

  7. Better accounting/charging/account management procedures: 8 responses
    • 3 - provide better accounting program; setcub hard to use or confusing
    • 2 - account creation too slow (1 was for HPSS account)
    • 1 - setcub scheme not enforced automatically on the T3E
    • 1 - resource allocation should reflect user needs: talk to users in account before setting up the account
    • 1 - better account and queuing policies that allow a group to work as cohesive unit

    NERSC is working on a plan to redesign CUB, the database which manages accounting information. Deployment of the new system is expected for FY'2000. More account creation procedures will be automated this year, especially for HPSS accounts.

  8. Keep users better informed; better interactions with users: 6 responses
    • 3 - keep users better informed about system and status changes
    • 2 - be more user oriented (rather than computer systems oriented)
    • 2 - inform all users of changes during the allocation award process, not just PIs
    • 1 - improve consultant quality
    • 1 - keep users informed about your plans for the future (e.g. announce CFS to HPSS plans now)

    One of the things NERSC learned from the survey was that users would like more email notification from NERSC. We have already started sending out more information by email in response to this.

  9. Improve the T3E / other T3E issues: 6 responses
    • 2 - keep pierre separate
      "My guess it will lower down overall efficiency and realability."
    • 1 - resolve T3E performance issues
    • 1 - get ssh
    • 1 - make it easier to do development work
    • 1 - more flexibility in allocation of memory and inodes
  10. Faster networks / better response time: 5 responses
    • 4 - speedier network access
    • 1 - improve ftp speed on Killeen
  11. Less down time: 4 responses
    • 2 - for T3E/HPSS
      • "Twice-weekly mid-day T3E downtimes / HPSS downtimes were a nuisance."
    • 2 - less downtime in general   

     

  12. Provide a workstation farm; better PDSF support: 4 responses

       

  13. Visualization support/software: 4 responses
    • 1 - improve graphics support to the average user
    • 1 - improve access to the vis server
    • 1 - provide g-sharp or similar software
    • 1 - provide readily available 4-D graphics and animation
  14. Increase research / scientific computing support: 3 responses
    • Need more people in scientific computing.
    • "Have experts in areas of user interest available as resources."
    • "Add to research computing facilities to compliment the production facility. Explore innovative use of non-numerical hardware and software, e.g. visualization environments."
  15. No changes needed: 3 responses
  16. Miscellaneous: 3 responses
    • interoperability with other systems (workstations, etc.)
    • X server
    • "My principal aggravation is in being asked to provide research documents to NERSC..."

 

Q9. How does NERSC compare to other centers you have used?


  1. NERSC is best, better than; favorable evaluation: 31 responses
    • better help from consultants, consultants are truly experts
    • more user friendly, better user support, more responsive
    • excellent computer support and CS expertise
    • more computing power than [another] center, more and better resources
    • web manuals easy to use and understand, lots of template scripts and programs
    • more reliable and easier to use, a usable production facility

     

  2. NERSC is the same as; mixed evaluation; not bad: 11 responses
    • responded faster but not as accommodating for some things
    • best overall for user support but not as good at simply providing huge chunks of time
    • good in general consulting services but bad in usability of machines; bad queue structure and charging system on C90

     

  3. NERSC is less good: 5 responses
    • not good for code development; NERSC does not support new ideas

     

  4. No experience with other centers; N/A: 5 responses

     

  5. Provided description of other centers but no evaluation; no comment: 4 responses

 


Q10.   If you were in charge of NERSC what changes would you implement? 

(see Summary)


 


Q11.   How long have you used NERSC?

More than 3 years: 60
6 months - 3 years: 55
6 months or less: 18

    


Q12.  What desktop systems do you use to access NERSC?

SystemN
Unix 107  
Mac 34  
PC-Windows 25  
PC-NT 20  
PC-linux 20  
other 8  
PC-DOS 2  

    


Q13.   How do you access NERSC machines?

SystemN
telnet 127  
ftp 93  
ssh 16  
other 2  



Q14.   What access methods would you like to use in the future?

 

  • 19 - telnet
  • 16 - ssh
  • 11 - ftp
  • 5 - rlogin & other R commands (1 via Hummingbird software)
  • 2 - web
  • 1 - secure telnet
  • 1 - remonte X
  • 1 - xterm
  • 1 - CVS with ssh
  • 1 - pftp & dpss
  • 1 - requires support for very large transfers to/from remote sites

Q15.   How satisfied are you with the NERSC website?

TopicNAvg.
Getting Started web pages 76 5.54
NERSC website in general 106 5.54
T3E web info 75 5.48
Accounts management web pages 71 5.38
J90 web info 64 5.34
Online tutorials 78 5.17
HPSS web info 52 5.10
What's New web page 58 4.95
NERSC News (online magazine) 61 4.92
Research web pages 54 4.91
Software documentation on web 72 4.85
Online Cray manuals 71 4.85
Visualization web pages 42 4.76



Q16.   Comments about our web pages

 

NERSC responses are in bold.

7 in general, good
6 don't use (much)
5 batch info needs improvement;
want to understand queueing algorithm and policies;
interpret codes from qstat and cqstatl;
online tutorial for running batch confusing;
hard to find info on how to submit runs - like max memory sizes for different queues;
hard to find sample scripts)
3 hard to navigate and search;
have smaller, more manageable parts;
info needs to be better organized
[New website organization by December 1998; new search engine by January 1999]
2 NERSC home page graphic too big [ A new home page is coming soon.]
1 publicize website [ The welcome email we send to new users now does that.]
1 email whatsnew to all users [We will use email a lot more in FY99.]
1 provide better examples on multiple processor environment; show hot to clean environment upon termination
1 online tutorial for cft77 to f90 confusing
1 errors in C++ info
1 T3E status/downtime is posted too late
1 software pages need improving
1 hardware pages good
1 not useful
1 getting started pages good
1 info not always up-to-date
1 need revision dates [We are in the process of implementing these.]
1 Cray manuals take a long time to load

Q17.   How do you find out about changes at NERSC?

MethodN
email from NERSC 84
motd (message of the day) 67
What's New area of the web 36
from colleagues 36
NERSC News 18
Change pages on the web 10
phone calls from NERSC 7
other 4



Q18.   Do you feel you are adequately informed about NERSC changes? If not, why not?

 

Yes 94, No 21

If not, why not?

  • I don't usually know if there has been a system change that is not listed on the message of the day.


  • I don't log in very often


  • Don't login much, didn't know about the website.


  • If I don't log into the system some time or check the web, I usually miss some announcements


  • Because I rely only on log-in messages. I don't have a lot of extra time to check websites frequently for messages.


  • I miss the Buffer - the Web is not a complete replacement.


  • I ought to check your NERSC news regularly


  • I don't check the what's new or change pages regularly.


  • I don't check the website very frequently and notices via email don't appear very often.


  • Sometimes I don't know about changes until after they occur.


  • I did not know that the NQS script format has changed until I accidentally found that.


  • I've had problems with my email and didn't realize free time on Pierre was over, thus overrunning my allocation.


  • I am mostly logged in on one of the workstations around the lab, since I can't make use of the CRAYs I am seldom logged in there. However, news are posted only there. In addition it would be nice to send a reminder each time even though it's a scheduled maintenance at least to PDSF, which is a part of NERSC, right? This is mainly related to HPSS.

Q19.   How would you like to be kept informed?

 

32 email (of which 6 email + web; 1 email to PDSF users re maintenance)
8 motd (1 said there used to be a lot more)
2 automated What's New
2 from colleagues
1 online bulletin board accessible during telnet session
1 are there email discussion groups?
1 more advanced notification of when T3E will/won't have scheduled down
1 broadcast messages

Q20.   How satisfied are you with NERSC's consulting services?

TopicNAvg.
Overall quality of consulting responses 113 5.88
Problem resolution by consultants 109 5.73
Software bug resolution 84 5.61
Consultants follow-up to questions 99 5.57



Q21.   Comments about consulting support

24 satisfied
4 don't use / don't need
2 mixed evaluation
2 consultants can't fix things:
"I'll report it to Cray; "we don't have that software"; "there is no way to do that" rather than "You can't do that now, but we'll get our system software programmers to write a utility which will allow it to be done in the future"
1 not useful
1 get more consultants
1 sometimes emails get lost
1 want faster email responses
1 didn't get response on query for CERNLIB on T3E

 

    


Q22.   What is needed to adequately cover your training needs?

MethodN
Online tutorials 85
Web-broadcasted training 28
Classes given in Berkeley 15
Classes given at remote sites 12

    


Q23.   Have you taken any of our training classes?

Yes 19,   No 102


Q24.   Comments about our online tutorials, suggested improvements and other ideas for training    

4 good (especially f90)
3 don't use / no opinion
1 provide test problems
1 give batch templates
1 maintain online FAQs
1 provide an interactive online training program
1 use advanced collaboration tools
1 classes provide no extra value over online tutorials
1 classes don't answer the questions you really want answered

 


Q25-27.   How satisfied are you with NERSC's large scale computers?

TopicNAvg.
C90 availability (uptime) 62 5.87
T3E interactive response time 66 5.74
J90 availability (uptime) 61 5.69
T3E uptime 72 5.58
C90 programming environment 59 5.58
T3E programming environment 69 5.51
J90 programming environment 55 5.40
J90 interactive response time 59 5.24
C90 interactive response time 61 5.21
J90 queue structure 55 4.85
C90 queue structure 52 4.85
J90 batch job turnaround time 57 4.79
T3E queue structure 67 4.51
T3E batch turnaround time 67 4.43
C90 batch job turnaround time 53 4.42



Q28.   Comments about NERSC's large scale computers suggested improvements, future needs

 

    


Q29-32.   How satisfied are you with NERSC's file storage systems?

TopicNAvg.
CFS reliability 52 5.87
CFS availability (uptime) 52 5.73
HPSS reliability 49 5.51
HPSS performance (bandwidth) 48 5.46
HPSS availability (uptime) 49 5.39
CFS performance (bandwidth) 51 5.37
HPSS response time 49 5.29
CFS response time 52 5.02
HPSS user interface 48 4.88
File restores from backup 46 4.83
CFS user interface 51 4.82
Data migration 56 4.80

 


 

Q33.   AFS

 

Do you use AFS on your desktop Yes 11 No 77
Is the AFS gateway dano.nersc.gov, as implemented on the J90 cluster and T3E, satisfactory? Yes 9 No 1
If NERSC migrates from AFS to DFS, would you use DFS at your site? Yes 4 No 2 Don't Know 37



Q34.   Comments about file systems, suggested improvements, future needs    

4 don't use mass storage (much)
2 want AFS
1 slow network traffic with AFS
1 miss masget/masput
1 CFS slow from T3E
1 faster restoration of migrated files
1 keep CFS data when move to HPSS
1 haven't tried HPSS; getting passwords too painful
1 if NERSC migrates to DFS, make mass storage part of DFS
1 little faith in them
1 no opinions
1 don't know what this is

 

 


 

 

Q35.   How satisfied are you with NERSC's visualization services?

TopicNAvg.
Escher availability (uptime) 14 5.07
Escher interactive response time 15 4.87
Escher software provided 16 4.81



Q36.   Comments about visualization services, suggested improvements, future needs

 

  • 12 - don't use
  • 1 - uses of vis server not clear
  • 1 - want something in between avs and gnuplot: like g-sharp
  • 1 - want full documentation & tutorials on all vis apps; especially avs, khoros, dx
  • 1 - don't need much
  • 1 - no strong opinions

Q37.   Anything else you want to say?

 

14 doing good job / satisfied
2 T3E too busy
2 longer / different queues on T3E
2 haven't used NERSC much
1 relative allocation of T3E resources between GC and non GC users is a difficult policy issue (from a non GC user)
1 all account users should be notified when increase in allocation not just PI
1 need petaflop computing
1 computer power misappropriated; should only be for what you can't do on desktop
1 consider SMP
1 keep NERSC reliable place for production computing
1 periodically remind us of your services
1 customer service good
1 setcub command inconvenient
1 Newton needs to de reviewed for sys admin procedures
1 HPSS needs improvement