Installing SPEC MPI2007 Under Unix and Linux

Last updated: 22 October 2009 cp
(To check for possible updates to this document, please see http://www.spec.org/mpi2007/Docs/ )


Contents


Installation Steps

The SPEC MPI2007 suite has been tested under Unix and Linux systems; the installation procedures described here are expected to work under Mac OS X, as well. Your DVD can be installed under many operating systems.

Reminder: the SPEC license allows you to install on multiple systems as you may wish within your institution; but you may not share the software with the public.

The installation procedure for Unix, Linux, and Mac OS X is as follows:

  1. Review the hardware and software requirements, in system-requirements.html

    Note: links to SPEC MPI2007 documents on this web page assume that you are reading the page from a directory that also contains the other SPEC MPI2007 documents. If by some chance you are reading this web page from a location where the links do not work, try accessing the referenced documents at one of the following locations:

  2. Create a directory on the destination disk. You should make sure that you have a disk that has at least the following free file system space:

    (For more information on disk usage, see system-requirements.html.)

  3. Mount the DVD. For Mac OS X and some versions of Unix and Linux such as SUSE the DVD will be automatically mounted. For some systems, you may have to enter an explicit mount command. If your operating system supports the Rock Ridge Interchange Protocol extensions to ISO 9660, be sure to select them, unless they are the default. The following examples are not intended to be comprehensive, but may get you started or at least give you clues which manpages to read:

    AIX: mount -v cdrfs -r /dev/cd0 /cdrom
    HP-UX: pfs_mountd &
    pfsd &
    pfs_mount -v /dev/rdsk/c1t2d0 /cdrom

    (Obtain the device name via ioscan -funCdisk. See the manpages for pfs, pfsd, pfs_mount, pfs_mountd, and pfs_umount. You will need the latter in order to eject the DVD.)
    Linux: mount -t iso9660 -o ro,exec /dev/cdrom /mnt
    Solaris: If Volume Management is running, you should find that the DVD is automatically mounted, as /cdrom/label_of_volume/ If not, you should be able to mount it with commands similar to this:
    mkdir /mnt1
    mount -F hsfs -o ro /dev/dsk/c0t6d0s0 /mnt1

    Note that you may need root privileges to mount the DVD.

    The following paragraphs assume that your DVD drive is on the same system as where you wish to install. If it is on a different system, please see the appendix.

  4. Set your current working directory to the directory where the DVD is mounted.

  5. Type
        ./install.sh
    When prompted, enter the destination directory. Note that for a destination directory, you will have to use syntax acceptable to sh (so you might need to say $HOME instead of ~).

    The installation procedure will attempt to determine your current architecture and install the appropriate tools. It will present you with a list of possible choices following this message:

          The following toolsets are expected to work on your platform.

    If you do not want to use the toolset that was provided, use the "-e" switch on install.sh to exclude those tools that are to be avoided.

    If the installation procedure is unable to determine your system architecture, you will see a message such as this one:

           We do not appear to have vendor supplied binaries for your architecture.  

    If you see that message, please stop here, and examine the file tools-build.html

    Thousands of files will be installed on your destination disk. (If you would prefer to see them all named you can set VERBOSE=1 in your environment before installing the kit.) You should see summary messages such as these:

        =================================================================
        Attempting to install the aix5L-ppc toolset...
    
        Checking the integrity of your source tree...
    
        Checksums are all okay.
    
    	Unpacking binary tools for linux-rhas4r4-ia64...  <<-- your valid architecture
    
        Checking the integrity of your binary tools...
    
        Checksums are all okay.
    
        Testing the tools installation (this may take a minute)
    
        ..........................................................o....................................o..........................................................................................o..
        All tests passed!
    
        Installation successful.  Source the shrc or cshrc in
    	'/scratch/tester/mpi2007/'          <<-- your directory
        to set up your environment for the benchmark.
    
    

    At this point, you will have consumed about 7GB of disk space on the destination drive.

  6. Change your current directory to the top-level SPEC directory and source either shrc or cshrc:

    The effect of the above commands is to set up environment variables and paths for SPEC.

  7. Test that you can build a benchmark using the config file supplied for your system. For example:

           % runspec --config=sgi-linux-ipf-intel-sgimpi.cfg  --action=build --tune=base --size=mref 137.lu 

    The above command assumes that you can identify a config file (in the directory $SPEC/config) that is appropriate for you. Note that in this and subsequent examples we have copied the example-sgi-linux-ipf-intel-sgimpi.cfg file and renamed it sgi-linux-ipf-intel-sgimpi.cfg. If you were to use the original filename, you would see the following warning every time you invoke runspec:

            =============================================================================
            Warning:  You appear to be using one of the config files that is supplied
            with the SPEC MPI2007 distribution.  This can be a fine way to get started.
    
            Each config file was developed for a specific combination of compiler / OS /
            hardware.  If your platform uses different versions of the software or
            hardware listed, or operates in a different mode (e.g. 32- vs. 64-bit mode),
            there is the possibility that this configuration file may not work as-is. If
            problems arise please see the technical support file at
    
              http://www.spec.org/mpi2007/Docs/techsupport.html
    
            A more recent config file for your platform may be among result submissions at
    
              http://www.spec.org/mpi2007/ 
    
            Generally, issues with compilation should be directed to the compiler vendor.
            You can find hints about how to debug problems by looking at the section on
            "Troubleshooting" in
              http://www.spec.org/mpi2007/Docs/config.html
    
            This warning will go away if you rename your config file to something other
            than one of the names of the presupplied config files.
    
            ==================== The run will continue in 30 seconds ====================
       

    For a quick introduction to config files, see "About Config Files" in runspec.html

    The "--tune=base" above indicates that we want to use only the simple tuning, if more than one kind of tuning is supplied in the config file.

  8. Test that you can run a benchmark, using the minimal input set - the "mtest" workload. For example:

           % runspec --config=sgi-linux-ipf-intel-sgimpi.cfg --size=mtest --tune=base \
                        --ranks=16 --noreportable --iterations=1 137.lu

    The "\" above indicates that the command is continued on the next line. The "--noreportable" ensures that the tools will allow us to run just a single benchmark instead of the whole suite, and "--iterations=1" says just run the benchmark once.

  9. Test that you can run a benchmark using the real input set - the "reference" workload. For example:

           % runspec --config=sgi-linux-ipf-intel-sgimpi.cfg --size=mref \
                        --ranks=16 --noreportable --iterations=1 137.lu

    Check the results in $SPEC/result.

  10. Have a look at runspec.html to learn how to do a full run of the suite. Also the command runspec -h will list the various options for runspec

Example Installation

Here is a complete installation on SGI Itanium/Linux systems, with interspersed commentary. This example follows the steps listed above. We assume that Steps 1 through 3 are already complete (from the list above).

Step 4: Set the current working directory to the DVD:

% cd /mnt/dvd/ 

Step 5: Invoke install.sh. When prompted, we enter the destination directory, and the desired architecture:

% ./install.sh

SPEC MPI2007 Installation

Top of the MPI2007 tree is '/mnt/dvd'
Enter the directory you wish to install to (e.g. /usr/mpi2007)
/scratch/tester/mpi2007

Installing FROM /mnt/dvd
Installing TO /scratch/tester/mpi2007

Is this correct? (Please enter 'yes' or 'no') 
yes

The following toolsets are expected to work on your platform.  If the
automatically installed one does not work, please re-run install.sh and
exclude that toolset using the '-e' switch.

The toolset selected will not affect your benchmark scores.

linux-rhas4r4-ia64            For Red Hat Enterprise Linux AS 4 or later.
                              Built on RHAS 4 update 4 with glibc-2.3.4-2.25

sles9-glibc23-ia64            For IA64 systems running SuSE Enterprise Linux 9
                              or later.
                              Built on SLES 9 with gcc 3.3.3.

linux-suse101-i386            For 32-bit x86 Linux systems running SuSE
                              Linux 10.1 or later.
                              Built on SuSE Linux 10.1 with 
                              gcc V4.1.0 (SUSE Linux)

linux-redhat62-ia32           For x86, IA-64, EM64T, and AMD64-based Linux
                              systems with GLIBC 2.1.3+.
                              Built on RedHat 6.2 (x86) with gcc 2.95.3



=================================================================
Attempting to install the aix5L-ppc toolset...

Checking the integrity of your source tree...

Checksums are all okay.

Unpacking binary tools for linux-rhas4r4-ia64

Checking the integrity of your binary tools...

Checksums are all okay.

Testing the tools installation (this may take a minute)

..........................................................o....................................o..........................................................................................o..
All tests passed!

Installation successful.  Source the shrc or cshrc in
/scratch/tester/mpi2007/
to set up your environment for the benchmark.

Step 6: Now, we set the current working directory to the DESTINATION device. If you are using a Bourne compatible shell, you will source shrc; if you are using a variant of the csh, you'll use cshrc.


$ cd /scratch/tester/mpi2007/ 
$ . ./shrc

That's dot-space-dot-slash-shrc just above.

Step 7: Use runspec to do a test build:


% runspec --config=sgi-linux-ipf-intel-sgimpi.cfg --action=build --tune=base --size=mref 137.lu
runspec v1292 - Copyright 1999-2009 Standard Performance Evaluation Corporation
Using 'linux-rhas4r4-ia64' tools
Reading MANIFEST... 10245 files
Loading runspec modules................
Locating benchmarks...found 19 benchmarks in 20 benchsets.
Reading config file '/scratch/tester/mpi2007/config/sgi-linux-ipf-intel-sgimpi.cfg'
Benchmarks selected: 137.lu
Compiling Binaries
  Building 137.lu base sgi default: (build_base_sgi.0000)

Build successes: 137.lu(base)

Build Complete

The log for this run is in /scratch/tester/mpi2007/result/MPI2007.002.log

runspec finished at Fri Sep  5 22:01:44 2008; 40 total seconds elapsed

Just above, various compile and link commands may or may not be echoed to your screen, depending on the settings in your config file. At this point, we've accomplished a lot. The SPEC tree is installed, and we have verified that a benchmark can be compiled using the C compiler.

Step 8: Now try running a benchmark, using the minimal mtest workload. The mtest workload runs in a tiny amount of time and does a minimal verification that the benchmark executable can at least start up. The number of ranks to use is set with the ranks option to runspec.

% runspec --config=sgi-linux-ipf-intel-sgimpi.cfg --ranks=16 --noreportable --tune=base --size=mtest --iterations=1 137.lu
runspec v1292 - Copyright 1999-2009 Standard Performance Evaluation Corporation
Using 'linux-rhas4r4-ia64' tools
Reading MANIFEST... 10245 files
Loading runspec modules................
Locating benchmarks...found 19 benchmarks in 20 benchsets.
Reading config file '/scratch/tester/mpi2007/config/sgi-linux-ipf-intel-sgimpi.cfg'
Benchmarks selected: 137.lu
Compiling Binaries
  Up to date 137.lu base sgi default


Setting Up Run Directories
  Setting up 137.lu mtest (test) base sgi default: created (run_base_mtest_sgi.0000)
Running Benchmarks
  Running 137.lu mtest (test) base sgi default
Success: 1x137.lu
Producing Raw Reports
mach: default
  ext: sgi
    size: mtest (test)
      set: medium
        format: raw -> /scratch/tester/mpi2007/result/MPIM2007.003.mtest.rsf

WARNING: There are no node descriptions present!  A fake one will be provided.

Parsing flags for 137.lu base: done
Doing flag reduction: done
        format: flags -> /scratch/tester/mpi2007/result/MPIM2007.003.mtest.flags.html
        format: ASCII -> /scratch/tester/mpi2007/result/MPIM2007.003.mtest.txt
        format: HTML -> /scratch/tester/mpi2007/result/MPIM2007.003.mtest.html, /scratch/tester/mpi2007/result/invalid.gif, /scratch/tester/mpi2007/result/MPIM2007.003.mtest.gif

The log for this run is in /scratch/tester/mpi2007/result/MPI2007.003.log

runspec finished at Sun Sep  7 19:47:25 2008; 22 total seconds elapsed

Notice about 20 lines up the notation "Success: 1x137.lu". That is what we want to see.

Step 9: let's try running 137.lu with the reportable mref workload. This will take a while - about 50 minutes on the 1.6GHz system used for the sample installation.

% runspec --config=sgi-linux-ipf-intel-sgimpi.cfg --ranks=16 --noreportable --tune=base --size=mref --iterations=1 137.lu
runspec v1292 - Copyright 1999-2009 Standard Performance Evaluation Corporation
Using 'linux-rhas4r4-ia64' tools
Reading MANIFEST... 10245 files
Loading runspec modules................
Locating benchmarks...found 19 benchmarks in 20 benchsets.
Reading config file '/scratch/tester/mpi2007/config/sgi-linux-ipf-intel-sgimpi.cfg'
Benchmarks selected: 137.lu
Compiling Binaries
  Up to date 137.lu base sgi default


Setting Up Run Directories
  Setting up 137.lu mref (ref) base sgi default: created (run_base_mref_sgi.0000)
Running Benchmarks
  Running 137.lu mref (ref) base sgi default
Success: 1x137.lu
Producing Raw Reports
mach: default
  ext: sgi
    size: mref (ref)
      set: medium
        format: raw -> /scratch/tester/mpi2007/result/MPIM2007.004.mref.rsf

WARNING: There are no node descriptions present!  A fake one will be provided.

Parsing flags for 137.lu base: done
Doing flag reduction: done
        format: flags -> /scratch/tester/mpi2007/result/MPIM2007.004.mref.flags.html
        format: ASCII -> /scratch/tester/mpi2007/result/MPIM2007.004.mref.txt
        format: HTML -> /scratch/tester/mpi2007/result/MPIM2007.004.mref.html, /scratch/tester/mpi2007/result/MPIM2007.004.mref.gif

The log for this run is in /scratch/tester/mpi2007/result/MPI2007.004.log

runspec finished at Sun Sep  7 20:43:37 2008; 2935 total seconds elapsed

Success with the real workload! So now let's look in the result directory and see what we find:

% cd result
% ls
MPI2007.002.log                MPIM2007.004.mref.flags.html
MPI2007.003.log                MPIM2007.004.mref.gif
MPI2007.004.log                MPIM2007.004.mref.html
MPIM2007.003.mtest.flags.html  MPIM2007.004.mref.rsf
MPIM2007.003.mtest.gif         MPIM2007.004.mref.txt
MPIM2007.003.mtest.html        invalid.gif
MPIM2007.003.mtest.rsf         lock.MPI2007
MPIM2007.003.mtest.txt

Notice the three separate sets of files: .002, .003, and .004

MPI2007.002.log has the log from the compile.

MPI2007.003.log has the log from running 137.lu with the "mtest" input. The various outputs (.html, .txt), a compiler flags report (.flags.html), and a raw output file (.rsf) are all preceded by "MPIM2007". The tools also distinguish the fact that the input was a "mtest" input by putting that in the file name as well.

MPI2007.004.log has the log from running 137.lu with the "mref" input. Once again, the various outputs all start with MPIM2007.

The careful reader may wonder: "What happened to *001* files?" The answer is that the author of this example had a fat-finger experience when trying to enter the correct runspec command and did not like the results, and so deleted them. Doing so is harmless to the other logs, and can help organize a result directory by getting the clutter out of the way. To remove everything associated with run #1, simply type "rm *001*"; or if you feel nervous about that, try something like "mkdir hideme; mv *001* hideme". Doing either of these will not cause the runspec tools to complain.

So let's look at our reference input results, in the .txt file:


% cat MPIM2007.004.mref.1.txt 
##############################################################################
#   INVALID RUN -- INVALID RUN -- INVALID RUN -- INVALID RUN -- INVALID RUN  #
#                                                                            #
# 'reportable' flag not set during run                                       #
# 137.lu (base) did not have enough runs!                                    #
# 132.zeusmp2 (base) did not have enough runs!                               #
# 126.lammps (base) did not have enough runs!                                #
# 127.wrf2 (base) did not have enough runs!                                  #
# 121.pop2 (base) did not have enough runs!                                  #
# 115.fds4 (base) did not have enough runs!                                  #
# 129.tera_tf (base) did not have enough runs!                               #
# 130.socorro (base) did not have enough runs!                               #
# 107.leslie3d (base) did not have enough runs!                              #
# 128.GAPgeofem (base) did not have enough runs!                             #
# 113.GemsFDTD (base) did not have enough runs!                              #
# 122.tachyon (base) did not have enough runs!                               #
# 104.milc (base) did not have enough runs!                                  #
# Unknown flags were used! See                                               #
#      http://www.spec.org/mpi2007/Docs/runspec.html#flagsurl                #
# for information about how to get rid of this error.                        #
#                                                                            #
#   INVALID RUN -- INVALID RUN -- INVALID RUN -- INVALID RUN -- INVALID RUN  #
##############################################################################
                           SPEC(R) MPIM2007 Summary
                                    -- --
                           Sun Sep  7 19:54:42 2008

MPI2007 License: --                                      Test date: Sep-2008
Test sponsor: --                             Hardware availability: --      
Tested by:    --                             Software availability: --      

                                  Estimated                       Estimated
                Base     Base       Base        Peak     Peak       Peak
Benchmarks      Ranks  Run Time     Ratio       Ranks  Run Time     Ratio
-------------- ------  ---------  ---------    ------  ---------  ---------
104.milc                                    NR                                 
107.leslie3d                                NR                                 
113.GemsFDTD                                NR                                 
115.fds4                                    NR                                 
121.pop2                                    NR                                 
122.tachyon                                 NR                                 
126.lammps                                  NR                                 
127.wrf2                                    NR                                 
128.GAPgeofem                               NR                                 
129.tera_tf                                 NR                                 
130.socorro                                 NR                                 
132.zeusmp2                                 NR                                 
137.lu             16       2920       1.26 *                                  
==============================================================================
104.milc                                    NR                                 
107.leslie3d                                NR                                 
113.GemsFDTD                                NR                                 
115.fds4                                    NR                                 
121.pop2                                    NR                                 
122.tachyon                                 NR                                 
126.lammps                                  NR                                 
127.wrf2                                    NR                                 
128.GAPgeofem                               NR                                 
129.tera_tf                                 NR                                 
130.socorro                                 NR                                 
132.zeusmp2                                 NR                                 
137.lu             16       2920       1.26 *                                  
 Est. SPECmpiM(TM)_base2007            1.26
 Est. SPECmpiM_peak2007                                             Not Run


                              BENCHMARK DETAILS
                              -----------------
      Type of System: Heterogenous Cluster
 Total Compute Nodes: 0
         Total Chips: 0
         Total Cores: 0
       Total Threads: 0
        Total Memory: 0 MB
      Base Ranks Run: 16
  Minimum Peak Ranks: --
  Maximum Peak Ranks: --
          C Compiler: --
        C++ Compiler: --
    Fortran Compiler: --
       Base Pointers: --
       Peak Pointers: Not Applicable
         MPI Library: --
      Other MPI Info: --
      Pre-processors: --
      Other Software: --

                      Node Description: Nonexistant Node
                      ==================================


                                   HARDWARE
                                   --------
     Number of nodes: --
    Uses of the node: --
              Vendor: --
               Model: --
            CPU Name: --
    CPU(s) orderable: --
       Chips enabled: --
       Cores enabled: --
      Cores per chip: --
    Threads per core: --
 CPU Characteristics: --
             CPU MHz: --
       Primary Cache: --
     Secondary Cache: --
            L3 Cache: --
         Other Cache: --
              Memory: --
      Disk Subsystem: --
      Other Hardware: --
             Adapter: --
  Number of Adapters: --
           Slot Type: --
           Data Rate: --
          Ports Used: --
   Interconnect Type: --


                                   SOFTWARE
                                   --------
             Adapter: --
      Adapter Driver: --
    Adapter Firmware: --
    Operating System: --
   Local File System: --
  Shared File System: --
        System State: --
      Other Software: --


                                General Notes
                                -------------
    I forgot to describe any nodes in my config file or raw file!


                                 Submit Notes
                                 ------------
    The config file option 'submit' was used.

                              Base Unknown Flags
                              ------------------
 137.lu: "ifort" (in FC) "ifort" (in LD) "-O2 -ftz" (in FOPTIMIZE)
         "-lmpi" (in EXTRA_LIBS)


    SPEC and SPECmpiM are registered trademarks or trademarks of the
    Standard Performance Evaluation Corporation.  All other brand and
    product names appearing in this result are trademarks or registered
    trademarks of their respective holders.
##############################################################################
#   INVALID RUN -- INVALID RUN -- INVALID RUN -- INVALID RUN -- INVALID RUN  #
#                                                                            #
# 'reportable' flag not set during run                                       #
# 137.lu (base) did not have enough runs!                                    #
# 132.zeusmp2 (base) did not have enough runs!                               #
# 126.lammps (base) did not have enough runs!                                #
# 127.wrf2 (base) did not have enough runs!                                  #
# 121.pop2 (base) did not have enough runs!                                  #
# 115.fds4 (base) did not have enough runs!                                  #
# 129.tera_tf (base) did not have enough runs!                               #
# 130.socorro (base) did not have enough runs!                               #
# 107.leslie3d (base) did not have enough runs!                              #
# 128.GAPgeofem (base) did not have enough runs!                             #
# 113.GemsFDTD (base) did not have enough runs!                              #
# 122.tachyon (base) did not have enough runs!                               #
# 104.milc (base) did not have enough runs!                                  #
# Unknown flags were used! See                                               #
#      http://www.spec.org/mpi2007/Docs/runspec.html#flagsurl                #
# for information about how to get rid of this error.                        #
#                                                                            #
#   INVALID RUN -- INVALID RUN -- INVALID RUN -- INVALID RUN -- INVALID RUN  #
##############################################################################
-----------------------------------------------------------------------------
For questions about this result, please contact the tester.
For other inquiries, please contact webmaster@spec.org.
Copyright 2006-2009 Standard Performance Evaluation Corporation
Tested with SPEC MPI2007 v72.
Report generated on Sun Sep  7 20:43:36 2008 by MPI2007 ASCII formatter v1041.

Done. The suite is installed, and we can run at least one benchmark for real (see the report of the time spent in 137.lu above).


Appendix: the DVD drive is on system A, but I want to install on system B. What do I do?

If the title of this section describes your situation, you basically have two choices.

  1. Network mount: You can mount the device over the network and do the installation remotely.
  2. Tar file: You can install from the tar file

1. Network mount

You might be able to mount the DVD on one system and use network services to make it available on other systems.

Please note that the SPEC MPI2007 license agreement does not allow you to post the DVD on any public server. If your institution has a SPEC MPI2007 license, then it's fine to post it on an internal server that is accessible only to members of your institution.

Whether you attempt a network mount will probably depend on:

If your network environment allows easy cross-system mounting, or if you feel brave about reading manpages, you can use a network mount for the installation. Otherwise, you can fall back on the tar file.

Network mount, easy:
for example, System A Solaris/Opteron  +  System B Solaris/SPARC

Your operating system may be configured to automatically mount the drive and automatically make it visible to other network systems, or may make it visible with minimal user intervention. During one set of testing, system A (with the DVD drive) was an Opteron-based system running Solaris 10. The SPEC MPI2007 DVD was inserted. The operating system mounted it automatically, and from a terminal window, a (non-privileged) user entered the Solaris share command to make it visible to other hosts.

On System B, a Solaris SPARC system, a non-privileged user typed:

cd /net/systemA/cdrom/spec_mpi2007
./install.sh

and the installation proceeded normally, picking up from step 5, above.

Network mount, medium difficulty:
for example, System A Solaris/Opteron  +  System B Tru64 Unix/Alpha

Subsequent to the tests of the previous paragraphs, the DVD drive on System A (Solaris/Opteron) was also visible to a system running Compaq Tru64 UNIX V5.1A. But in this case, a little assistance was needed from the privileged (root) account on system B:

echo "systemA.domain.com:/cdrom/spec_mpi2007 /systemA nfs ro,bg,soft 0 0" >> /etc/fstab
mkdir /systemA
/usr/sbin/mount /systemA

Then, the non-privileged user was able to say:

cd /systemA
./install.sh

and once again the installation proceeded normally, picking up from step 5, above.

Network mount, a bit harder:
for example, System A SuSE/x86  +  System B Mac OS X/PowerPC

The SPEC MPI2007 DVD was also inserted into a system running SuSE Linux 9.0, and used from a Mac OS X PowerBook. On both these systems, there are probably automatic tools that would have accomplished the following more quickly, but the tester happened to read the manpages in the particular order that he happened to read them in. The following succeeded:

On System A, root added

/dev/cdrom  /cd  iso9660  ro,user,noauto,unhide

to /etc/fstab as suggested by man mount; the DVD was inserted; and the user typed mount /cd. On System A, root also added:

/cd  192.168.0.0/24(ro,insecure,no_root_squash,sync)

to /etc/exports, and then typed:

exportfs -r
rpc.nfsd -p 8
rpc.mountd
cat /var/lib/nfs/etab

On System B, root typed:

mkdir /remote
mount -t nfs 192.168.0.106:/cd /remote

Finally, the user typed

cd /remote
./install.sh

and installation continued as normal, with step 5.

2. Tar file

If the DVD drive is on a system other than the one where you wish to do the installation, and if you do not wish to try to get a network mount working, then the final fallback is to use the compressed tarfile. If you choose this option, please carefully observe the warnings.

  1. Go to the system with the DVD drive ("System A"). Insert the SPEC MPI2007 DVD, and, if required, issue a mount command.

  2. From a terminal window (aka command window), cd to the top level directory on the DVD.

  3. You are going to retrieve five things from the DVD. First, find the large tarfile and its corresponding md5 file:

    cd install_archives 
    ls -l mpi2007-2.0.tar.bz*
    

    You should see one moderately large file > 400MB, mpi2007-2.0.tar.bz2, and a small file associated with it that contains a checksum, mpi2007-2.0.tar.bz2.md5.

    If you don't see the above files, try looking for mpi*tar*. The name might change if, for example, a maintenance update of MPI2007 changes the name slightly to indicate an updated version.

    Do whatever is required in order to transfer both files intact to the system where you wish to do the installation ("System B"). If you use ftp, do not forget to use image (binary) mode. For example:

    $ ftp
    ftp> op systemB
    Name: imauser
    Password:
    ftp> cd /kits
    ftp> bin   <-------- important
    200 Type set to I.
    ftp> put mpi2007-2.0.tar.bz2
    ftp> put mpi2007-2.0.tar.bz2.md5
    

    Please note that the SPEC MPI2007 license agreement does not allow you to post the above file on any public ftp server. If your institution has a SPEC MPI2007 license, then it's fine to post it on an internal server that is accessible only to members of your institution.

  4. Next, you are going to look on the DVD for versions of specbzip2, specmd5sum, and spectar that are compatible with system B. Please do not use the tar supplied by your operating system unless you are sure that it can handle long path names. Many commonly-supplied tar utilities cannnot.

    Please do not use Windows Zip utilities, as these will not preserve line endings.

    If you have GNU tar and the genuine bzip2, then you can use those; otherwise, please hunt around on the DVD to find prebuilt versions that are compatible with your environment, like so:

    $ cd /dvdrom
    $ cd tools
    $ cd bin
    $ ls
    aix5L-ppc          irix6.5-mips         linux-ydl23-ppc  sles9-glibc23-ia64
    freebsd54-ia32     linux-redhat62-ia32  macosx           solaris-sparc
    freebsd60-ia32     linux-rhas4r4-ia64   macosx-ppc       solaris-x86
    hpux-11iv1-parisc  linux-suse101-AMD64  netbsd21-ia32    tru64-alpha
    hpux-11iv2-ipf     linux-suse101-i386   openbsd38-ia32   windows-i386
    $ cd aix5L-ppc 
    $ cat description 
    For PowerPC systems running AIX 5L V5.1 or later
                                  Built on AIX 5L V5.1 with the
                                  IBM XL C/C++ for AIX Version 8 compiler
    $ ls -l spec*
    -r-xr-xr-x   1 root     root       83656 Feb 23  2007 specbzip2
    -r-xr-xr-x   1 root     root       25078 Feb 23  2007 specmd5sum
    -r-xr-xr-x   1 root     root      258102 Feb 23  2007 spectar
    $ 
    

    Once you've found the right versions of specbzip2, specmd5sum, and spectar for the system where you intend to install (system B), transfer them to system B using the same methods that you used for the big tarfile.

  5. On system B, use specmd5sum to check that the file transfer worked correctly. In this example, we assume that you have placed all 5 of the files mentioned above in the /kits directory:

    $ cd /kits
    $ chmod +x spec*
    $ specmd5sum -c mpi2007-2.0.tar.bz2.md5
    mpi2007-2.0.tar.bz2: OK
    
  6. Unpack the tarfile, like so:

    $ cd /mybigdisk
    $ mkdir mpi2007
    $ cd mpi2007
    $ /kits/specbzip2 -dc /kits/mpi2007-2.0.tar.bz2 | /kits/spectar -xf -
    

    Be patient: it will take a bit of time to unpack! It might take 15 minutes, depending on the speed of your processor and disks. Go for a coffee break.

  7. Now, at last, type ./install.sh and pick up with step 5, above. Your output will be similar, but not identical, to the output shown in step 5 above: you won't see the "Unpacking xxxx" messages, because you already did the unpacking.


Copyright © 1999-2010 Standard Performance Evaluation Corporation
All Rights Reserved