NERSC 1998 User Survey Results

These are the complete responses to this question, with a few minor edits.

Areas of concern:

• To often one gets the answer sorry, there is no way to do that". Seldom does one get the reply "That's how it is but we're working on changing it" or: "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".

• The consult staff is very eager to help and usually quick as well. Unfortunately the problem is often beyond their control, So the answer is "I'll report it to Cray", or to cite a specific example, "None of the existing sparse matrix routines are effective for nxn matrices with n of a few hundred". They clearly seem to be doing all they can with the existing resources, so the only obvious improvement is more resources. 8-)

• My only gripe (and it is only a MINOR one), would be that sometimes my e-mails have seemed to get lost in the shuffle somewhere.

• Would like faster response to questions via email.

• respond faster to our questions

• At the begin of our usage of NERSC, we ported to T3E the part of CERNLIB we needed. We sent a message to the consulting support, but we did not hear from them. CERNLIB is a package used widely in physics society. We wish in the future this kind of user activity can be acknowledged.

• some consultants are excellent, others are very lacking

• I don't know, mostly the consultants (note mostly, not always) look like they do not know anything or they know everything. The should be provided with some kind of ranking of each user to try to answer the questions at the user's level. I know that this is hard under all points of view, but if the something is done there would help.

• Seems not very useful, though my impression may be wrong.

• I was very pleased to see dedication.

• great bunch of people!

• Impressively competent and empathic.

• My experience with the consultants has been great! It seems they have ample resources to answer bone-headed queries as well as complicated, multi-faceted problems.

• None. If they don't know the answer, then they know who to direct me to. It works quite well.

• they help me a lot. I am very satisfied

• This is key part of what make NERSC so good.

• very supportive

• Very helpful

• maintain the high quality!

• I was impressed that the consultants were able to track down and report to Cray software a problem with the compiler optimization.

• More people?

• the NERSC consultants do an excellent job. Call backs and problem resolution are highly commendable.

• I like it very much

• Overall, very responsive and helpful.

• The consultants are very good. This is a big improvement over the operation at LLNL.

• Overall, I am very happy with the consulting services at NERSC. I think that this area has been the most improved since the move to Berkeley.

• User problems are taken seriously.

• Awesome

• I have gotten very good response and follow up on my questions.

• Tom DeBoni has been very helpful in questions concerning multiple PCU use on the J90's. Richard Gerber has also been quite helpful in a number of consulting areas.

• As mentioned above, I was very pleased with Harsh Anand and Tom Deboni, although others were helpful too.

• I have received excellent support from the NERSC consultants. I work particularly closely with Mark Durst but also receive assistance from the other consultants quite often.

• We have an ongoing connection with Mark Durst, who has been unfailingly efficient and helpful (and pleasant to work with, too!)

• There are enough knowledgeable people locally who are able to answer my questions, so I have never had to contact NERSC consultants directly.

• Never used any.

• Haven't really used the consulting services.

• I haven't requested any debugging help

Back to Survey Summary Page