ARSC HPC Users' Newsletter 225, August 3, 2001
- Call for Papers: ARSC IBM SP Workshop
- SV1 Memory Banks and Bandwidth
- ARSC Student Intern Presentations, Aug 9
- ARSC Faculty Camp: Lectures Open to Local Folks
- Quick Tip
Call for Papers: ARSC IBM SP Workshop
ARSC Workshop : CALL FOR PAPERS
Application Development and Tuning Workshop
September 4-7, 2001
Arctic Region Supercomputing Center Fairbanks Alaska
The Arctic Region Supercomputing Center (ARSC) will host a 4-day workshop on IBM SP system use, benchmarking methodologies, and related topics. In the mornings, users will describe their experiences developing and tuning scientific applications for various high performance computing systems and discuss the requirements of some emerging computational challenges. In the afternoons, staff from the IBM Advanced Computing Technology Center (ACTC) will provide IBM technology updates and training on selected topics to help users get the most out of SP systems.
This workshop is open to all members of the DoD HPCMP and University of Alaska communities. System access will be limited to users who have applied for, or hold, a National Agency Check.
A dedication event for ARSC's new IBM SP system will be held.
Call for Papers
To present a paper at the workshop, please submit a brief (no more that 150 words) abstract by August 15, 2001. Each presentation should be 30 minutes including 5 minutes for Q&A.
Presentations will be selected for their relevance to the HPC community. Notification will be August 17, 2001. Topics of interest include:
- Benchmarking of Application Codes
- Application Tuning & Optimization
- Computational Requirements for Emerging Technologies
Submit abstracts as plain text or MS Word attachments to Guy Robinson via email: email@example.com
The workshop will be conducted in the afternoon session and will be specific to the IBM SP architecture. The sessions will focus around the following general topics:
- Compilers and Programming Environments
- MPI and OpenMP
- Tools for Optimization
- Q and A and additional hands-on
Additional information will be posted on the ARSC and HPCMP web pages as it becomes available:
for more on streaming.) A T3E run with streaming turned off increased the time from 0.340 to 0.816 seconds.
ARSC Student Intern Presentations, Aug 9
Beginning at 2pm, Aug. 9, in Butrovich 109, ARSC's three summer interns will be presenting their projects over the AGN. These talks are open to interested members of the ARSC/UAF community, and, of course, to those with an Access Grid Node:
Annemarie Dahm (University of Montana) will present her work augmenting the GSLib kriging algorithms library to allow for interpolation of precipitation and other hydrologic data.
Orlando Solis (University of Texas El Paso) will present his system for automating and organizing regression testing across multiple platforms.
Doug Vicere (University of Alaska Anchorage) will present his work on a tsunami model, generalizing input and other aspects of the code, and producing a vizualization tool to help in code verification.
ARSC Faculty Camp: Lectures Open to Local Folks
Limited space is available for any interested users (or potential users) to attend the following lectures in ARSC's 2001 "Faculty Camp." Contact Guy Robinson, 474-6386, firstname.lastname@example.org if you are interested, and he'll provide details.
First Week. August 6-10 ================================= Monday 10am: Introduction to ARSC Tuesday 10am: Users Introduction to ARSC Supercomputers 2pm: Introduction to Parallel Processing Second Week. August 13-17 ================================= Tuesday 10am: Cray Presentation on Computing Technologies Thursday/Friday Workshop on Creating Computational Scientists in the 21st Century Third Week. August 20-24 ================================= Tuesday 10am: Visualization Case Studies. Thursday 2pm: Presentations and Faculty Camp achievements
If you can't attend, ARSC consultants and specialists are available for one-on-one discussions and will be putting together ARSC's fall 2001 schedule of courses soon. Let us know where your interests lie!
Quick-Tip Q & A
A:[[ I had some .lst and .lst.gz files, all of which I wanted to [[ hide in a separate subdirectory. I used "mv" (the cryptic Unix [[ alternative, it seems, to click and drag). It seems to have worked, [[ but "mv" complains, and I don't know if I should worry about this. [[ [[ termite$ ll [[ total 10 [[ drwx------ 2 morty groop 4096 Jul 12 17:21 lst_files/ [[ -rw------- 1 morty groop 34129 Jul 12 17:20 ptsv.f [[ -rw------- 1 morty groop 20011 Jul 12 17:20 ptsv_gen.f [[ -rw------- 1 morty groop 791991 Jul 12 17:20 ptsv_gen.lst.gz [[ -rw------- 1 morty groop 10233 Jul 12 17:20 ptsv.lst.gz [[ -rw------- 1 morty groop 550 Jul 12 11:10 vgen.lst [[ termite$ [[ termite$ mv *lst* lst_files [[ lst_files - Invalid argument [[ termite$ [[ termite$ ll [[ total 8 [[ drwx------ 2 morty groop 4096 Jul 12 17:21 lst_files/ [[ -rw------- 1 morty groop 34129 Jul 12 17:20 ptsv.f [[ -rw------- 1 morty groop 20011 Jul 12 17:20 ptsv_gen.f [[ termite$ [[ termite$ ll lst_files [[ total 2 [[ -rw------- 1 morty groop 791991 Jul 12 17:20 ptsv_gen.lst.gz [[ -rw------- 1 morty groop 10233 Jul 12 17:20 ptsv.lst.gz [[ -rw------- 1 morty groop 550 Jul 12 11:10 vgen.lst [[ termite$ # # Three answers and a remark, sorted by verbosity: # # # Nic Brummell # It is trying to move the directory lst_files into itself. Fortunately, Unix does not allow this Escher-like activity and therefore complains. # # Guy Robinson... (editorial remark) # Actually it is safe on most modern unix systems. However, back when I was working on some of the first Unix ports, one user managed to issue a similar command, and the system very happily kept moving a into a/a into a/a/a into a/a/a/a until we were out of inodes. # # Richard Griswold # Since the regex *lst* matches the directory name lst_files, mv thinks you want to move lst_files into lst_files. The error, which varies from system to system, is telling you that you cannot move a directory into itself. You can avoid this error message by using a more fine-tuned regex, such as this: mv *.lst *.lst.gz lst_files However, as long as the files are moved correctly you don't have to worry. # # Bradford Chamberlain # In this command, what's happening is your use of wildcard characters is not only matching the .lst.gz and .lst files you targeted, but also lst_files (where the first * is evaluating to nothing and the second is evaluating to "_files"). Thus, the UNIX command thinks that you want to mv lst_files to lst_files, which is nonsensical, so it complains. Thus, the short answer is that this isn't something to worry about. It's just warning you that it's not doing the nonsensical thing you requested. If it seems like mv ought to be smart enough to know that this isn't what you meant by the wildcards, keep in mind that your shell is the culprit that's expanding the wildcard characters and sending the results to mv. mv is quite ignorant about the fact that you typed wildcard characters at all, as it's just handed the following command by the shell: mv lst_files ptsv_gen.lst.gz ptsv.lst.gz vgen.lst lst_files (which specifies to move all the named files but the last into the directory named by the last argument). My best suggestion for making this work is to use a better wildcard filter that doesn't match the directory name. For example, in your case you have the easy option of using: mv *.lst* lst_files which will not match the lst_files directory. If you really want to be a geek, you can use: mv ?*lst* lst_files which means "match anything that has at least one character followed by "lst" followed by any other character." Since your directory starts with "lst" this will cause it not to match. Q: I am rather upset, so please be nice. My Fortran program has variable called "sec", and I need to find wherever I subtract it. The coding is not always consistent, so these statements can take various forms, like: dtime=tstart-sec dtime = tstart - sec I figured I'd be clever, so I tried egrep, but it didn't work. So I tried grep, and got the same wierd output! It's a total miss: morty$ egrep "-sec - sec - sec" littleprog.F tail count(4) = 1 write(*,*) 'parameters for ncvgtg call' write(*,*) 'ncid ',ncid write(*,*) 'sec: ',sec write(*,*) 'Calling ncvgtg' call ncvgt(ncid,dataid,start,count,data,error) write(*,*) 'Finished calling ncvgtg' call ncclos(ncid,error) close(1) close(7) morty$ grep "-sec" littleprog.F tail count(4) = 1 write(*,*) 'parameters for ncvgtg call' write(*,*) 'ncid ',ncid write(*,*) 'sec: ',sec write(*,*) 'Calling ncvgtg' call ncvgt(ncid,dataid,start,count,data,error) write(*,*) 'Finished calling ncvgtg' call ncclos(ncid,error) close(1) close(7) So my question is: WHAT HAVE YOU CHANGED ON YOUR SYSTEM! (or... ahem, am I doing something wrong?)
[[ Answers, Questions, and Tips Graciously Accepted ]]
Ed Kornkven ARSC HPC Specialist ph: 907-450-8669 Kate Hedstrom ARSC Oceanographic Specialist ph: 907-450-8678 Arctic Region Supercomputing Center University of Alaska Fairbanks PO Box 756020 Fairbanks AK 99775-6020
Subscribe to (or unsubscribe from) the e-mail edition of the
ARSC HPC Users' Newsletter.
Back issues of the ASCII e-mail edition of the ARSC T3D/T3E/HPC Users' Newsletter are available by request. Please contact the editors.