By: Bill Buros. I thought this would be a relatively trivial exercise, but as with any good technology answer, it depends. It depends on your server, how it's controlled, and the distro releases you have. In this case, I have a Power7 server connected to an HMC with RHEL 6.1 installed on the server which is connected to a previously enabled RHEL 6.1 yum repo.
NOTE: That "enabled RHEL 6 yum repo" is an important assumption. You can check that you have that enabled with the command
# yum repolist
and look for the base "Red Hat Enterprise Linux 6Server - ppc64" repo.
The second note is these instructions assume you can get through company firewall's to get to outside repositories. You will require firewall access to public.dhe.ibm.com (22.214.171.124) and linuxpatch.ncsa.uiuc.edu (126.96.36.199)
Here we provide the short-cut links and steps for this process. Read the instructions if you have a different setup.
First, we highly recommend the new yum repo process. So to start, go to
and under the Yum repository section at the top of the page, click on the Learn more and download tools. Course, you could skip a step and just click on that "learn more" link here.
From the yum page, there a Configuration RPM file that you need to download and install on your system. In my case and at this time, the file I downloaded was named ibm-power-repo-1.1.6-5.ppc.rpm. I believe that this file works across RHEL 5, RHEL 6, SLES 10, and SLES 11, but here I'm just playing around with RHEL 6.1 so that's all I tested.
Step 1: rpm -i ibm-power-1.1.6-5.ppc.rpm
Once that's installed, you should install the recommended packages
Step 2: yum install ibm-power-managed-rhel6
In my case, the results of this command were:
sqlite.ppc 0:3.6.20-1.el6 src.ppc 0:188.8.131.52-11277
Finally, we very often take advantage of the latest Advance Toolchain compilers and optimized system libraries, so the last step for now is to install the latest Advance Toolchain 5.0 packages.
Step 3: yum install advance-toolchain-at5*
Which nicely downloads and installs the -devel, -perf, and -runtime rpms on the server. The Advance Toolchain packages are about 650MB in total, so that's something I usually start and return to later.
There are of course more packages, more recommendations, more tools. But this should get you started. You might consider the IBM SDK. We always recommend oprofile on your system.
By: Bill Buros.
Recently, the IBM SDK for Linux on Power released another beta update of the environment for developing, debugging, and doing performance analysis of programs running on Power systems. On a regular basis, people ask about the SDK and wonder "So... what exactly can this do?". Being a little tired of creating more and more wiki pages, I'm hoping that the video approach might be more interesting and quickly informative (not to mention it provided a nice tangent on the few days before the Thanksgiving break).
So here's a quick example of running a simple program (stream) in a 8-core partition on a Power7 server running SLES 11 sp1. "stream" measures memory bandwidth on a server running four different variants of matrix computations - essentially manipulating three matrices loaded in memory.
The videoclip is a shockwave clip, so I'm also testing to see how well this works on the Think Power Linux Community. With the software I'm using, the other alternative was mpeg-4 output, which I wanted to avoid here. Feel free to add your comments below - good - bad - indifferent.
Here's the various caveats: The IBM SDK was installed on a fresh SLES 11 sp1 partition taking advantage of the new YUM repo. That yum repo is VERY nice, check it out. In addition, oprofile was also installed and enabled in the SDK (a surprising manual step). The stream source can be downloaded from http://www.cs.virginia.edu/stream/FTP/Code/. We modified the stream makefile to take advantage of the OMP pragma's in the code for openmp parallelization. The example is loosely based on an older wiki page on developerWorks which focused on "tuning stream with libhugetlbfs".
Your performance results will of course vary, the example was run on a partitioned Power7 server.
Assuming this video approach works and is reasonable effective, there are some really nice tools and approaches being developed in and around the SDK which we'll elaborate on over the coming weeks. I will observe that one of the key challenges with the SDK will be in understanding how to import / leverage the far more complex code repositories and applications which customers and analysts often leverage. That process of importing codes is beyond my scope and interest level, so I'll leave that to the SDK experts to explain.
By: Jessica Erber-Stark.
Need a quick reference for setting up your Power Linux system? See Getting started with Linux on Power Systems Servers. After your system is up and running, use the IBM Installation Toolkit Simplified Setup Tool to easily configure open source workloads.
Many of our Power Linux users also asked for a concise guide about setting up virtualization on their system. The Getting started with virtualization using PowerVM topic provides an overview of the roadmap for virtualizing your system, specifics about the differences between the Power and x86 environments, and step-by-step instructions for configuring partitions.
New to the Power Linux information in the Linux Info Center, we are also including translated versions of the documentation for those who prefer a language other than English. The documentation linked here is available in the following languages:
By: Wainer dos Santos Moschetta.
It was released the version 0.8 of IBM SDK for Linux on POWER with several improvements:
The IBM SDK for Linux on POWER is a
By: Robert MacFarlan.
IBM, in collaboration with the Open Source Lab at Oregon State University, has contributed access to a new POWER7 server for use by the GCC development community. Quoting the email list announcement the configuration looks like:
"The server is an IBM Power 730 Express server configured with 16 POWER7 cores each with 4 hardware threads at 3.55 GHz, 64 GB of RAM and a few TB of disk. It is running Fedora Core 16 alpha."
Check out the recent blog posting on Fedora 16 Alpha for ppc64.
The GCC Community announcement can be found at: https://mail.gna.org/public/gcc-cfarm-users/2011-11/msg00000.html
By: Brent Baude.
I just wanted to bring some attention to the recent announcement around a Fedora 16 Alpha for ppc64. Detials can be found in the release announcement. While this is only an Alpha, it has proven to be a rather decent effort and quite stable.
For those with ppc64 hardware, I would
encourage you to check out the release notes for the alpha. This should give you some insight into any problems we have
already identified. To find out where to download the Fedora 16
alpha ISOs, check out the https://mirrors.fedoraproject.org/publiclist/Fedora/16-Alpha/ppc64/.
This is a community-based effort and as such we encourage folks to participate. The best mechanism to speak with the team behind this effort is via IRC, where we hang out on freenode in #fedora-ppc.
For those interested, the default userland is 64-bit. It can also run 32-bit binaries and yum (the “wrapper” admin tool around rpm) does an excellent job of resolving dependencies for package installation. Suppose you want to install a package called “foo”, you can perform any of the following:
Fedora, and this alpha, is a a great way to see trends in the Linux community operating systems (for which the Enterprise Linux distributions generally follow). One example is the incorporation of things like systemd, the new style of System and Service manager. If you have packages with init daemons and the like, running things like this alpha will help you get ready for full support of systemd.
I'll continue to post more information on Fedora for ppc64 in the future. My hope is to not only bring status but perhaps introduce new functions, tips, and techniques for users of Fedora.