Power Up!
The PowerStation comes with Yellow Dog Linux 6 pre-installed. YDL began life in 1999 as an alternative to the Mac OS on Apple's PowerPC hardware. It is based on Red Hat Enterprise Linux and Fedora. Like every Red Hat/Fedora derivative I have ever used, it uses RPM for package management. Yum, which started as a Yellow Dog-specific add-on for simplifying package updates (and has been adopted by most RPM-based distributions) is naturally included, along with the graphical yum updater, Pup.
When first booting the PowerStation, you go through the normal Anaconda new-user setup. Unfortunately, this process crashed on me at the very end, forcing me to reboot the computer. It seemed like it was just a fluke, so I didn't worry about it too much, but I think it may have contributed to my Firefox problems (more on that later).
The package selection available in the default YDL repositories is decent, but it's not as large as I am used to on Ubuntu and Fedora. The repositories configured out of the box include the Yellow Dog Base, Extras and Updates repositories, along with a PowerStation-specific one. Most of the basic apps—from Firefox to OpenOffice.org to The GIMP to PostgreSQL to Pidgin—are present and accounted for.
Although most of the packages I expected to find were available, a couple interesting ones were absent. One in particular (which happens to be one of my favorite applications), Inkscape, was missing. Thanks to the PowerPC Fedora Extras repository, I was able to install it easily.
A couple packages I wanted to use on the box, such as the renameutils (from www.nongnu.org/renameutils), were not available as pre-built RPM packages (as far as I could see, anyway). In the case of renameutils, I was able to download, compile and install the package manually.
One big thing I had to get used to on the PowerStation was the lack of GRUB. Yaboot is the bootloader for the PowerStation.
Being unfamiliar with Yaboot, I elected not to tinker with it or even spend much time looking at it. The system booted fine, and I didn't want to render the box unbootable inadvertently. The Yaboot configuration does look marginally similar to GRUB's, and I'll leave it at that.
Enlightenment is the default desktop environment for the PowerStation. The default theme and layout are nice, and the menus are well organized.
One interesting decision Terra Soft made was to configure the first Enlightenment desktop to run Nautilus full screen for file management purposes. This was a little confusing at first when I tried to change the Enlightenment wallpaper and nothing happened. To change the wallpaper of the first desktop, I had to change the GNOME wallpaper. Wallpaper issues aside, using Nautilus is a smart move, because it is one of the most advanced file managers available. I soon got used to using the first desktop for file management and the rest for running apps. This arrangement actually forced me to become better organized as different tasks were more clearly and cleanly separated. The other three desktops do not have Nautilus running on them, so they behave like regular Enlightenment desktops, animated backgrounds and all.
GNOME also is available out of the box as a session login option for those who prefer it, and KDE can be installed with the package manager.
There's no difference in running apps like The GIMP, OpenOffice.org or Firefox on an IBM Power processor as opposed to an Intel or AMD processor, so I won't go into running them other than to say they ran fine.
Performance is one of those tricky areas that are hard to define and nail down—especially when trying to compare the PowerStation to x86 workstations. One thing I can say without any equivocation is that the PowerStation definitely is speedy. Applications launch instantly (or nearly so), and everything feels fast and snappy. But “feelings” sometimes can lie, so to get a more accurate view, I turned to some performance testing.
For testing, I installed the Phoronix test suite. Unfortunately, although I was able to install it without trouble by following the directions on the Phoronix Test Suite Web site and run most of the tests, a few of them, including compiling the Linux kernel and calculating Pi to 32 million digits, failed. In the case of the Linux kernel compilation test, Phoronix reported that the test completed in 4.12 seconds. This compares to a time on my laptop of 4,407.53 seconds. Now, I am the first to admit that the PowerStation is much faster than my old laptop, but it is not a thousand times faster.
In the case of the Pi calculation test, the issue was that the test assumes you are running on x86, and it tries to load an x86 binary, which obviously won't run on the IBM 970MP-powered PowerStation. Thankfully, other tests in the Phoronix test suite provided more trustworthy results. The mencoder test, for example, which converts an 89MB avi file, took a respectable average of 42.13 seconds. See Table 1 for the results of some of the other Phoronix test suite tests I ran on the PowerStation.
Table 1. Phoronix Test Results on the PowerStation
| Test | Average Result |
|---|---|
| RAMspeed Average Integer Test | 2,661.78MB/s |
| OpenSSL | 36 signs per second |
| GnuPG 1GB file encryption | 21.96 seconds |
| Compress a 128MB file with lzma | 315.53 seconds |
| Compress a 512MB file with gzip | 39.38 seconds |
| Compress a 512MB file with parallel-bzip2 | 67.60 seconds |
| SQLite 2,500 insertions on indexed db | 74.05 seconds |
| SciMark composite test | 264.64Mflops |
| Timed PHP compilation | 82.93 seconds |
| IOMeter file server access pattern | 89.87 seconds |
| Timed Apache build | 54.48 seconds |
One caveat in my testing is that all of the tests shown in Table 1 finished with minor errors, such as “PHP Notice: Undefined offset: 0 in /usr/share/phoronix-test-suite/pts-core/functions/pts-functions_system_cpu.php on line X”. I don't think the errors skewed the results in either direction, but there is a bug somewhere that does not exist when the tests are run on an x86-based machine.
My original plan was to run these tests head to head against a quad-core Xeon system I was testing at the time. However, due to delays in receiving the PowerStation and some unforeseen issues with the Xeon system, I was not able to do this. Feel free to run the same tests on your workstation and compare your performance to the PowerStation.
One curious thing about the test results is that some of them are slower than the score achieved by my laptop—for example, the SQLite test. My old laptop is able to complete that test in an average of 62.63 seconds—more than ten seconds faster than the PowerStation. This may be because of the way the SQLite test works, or maybe it favors Intel processors—I don't know. In most tests, the PowerStation was faster, and in some tests significantly faster, which is what I expected.
Realizing the promise of Apache® Hadoop® requires the effective deployment of compute, memory, storage and networking to achieve optimal results. With its flexibility and multitude of options, it is easy to over or under provision the server infrastructure, resulting in poor performance and high TCO. Join us for an in depth, technical discussion with industry experts from leading Hadoop and server companies who will provide insights into the key considerations for designing and deploying an optimal Hadoop cluster.
Sponsored by AMD
Built-in forensics, incident response, and security with Red Hat Enterprise Linux 6
Every security policy provides guidance and requirements for ensuring adequate protection of information and data, as well as high-level technical and administrative security requirements for a system in a given environment. Traditionally, providing security for a system focuses on the confidentiality of the information on it. However, protecting the data integrity and system and data availability is just as important. For example, when processing United States intelligence information, there are three attributes that require protection: confidentiality, integrity, and availability.
Learn more about catching the bad guy in this free white paper.
Sponsored by DLT Solutions
| Designing Electronics with Linux | May 22, 2013 |
| Dynamic DNS—an Object Lesson in Problem Solving | May 21, 2013 |
| Using Salt Stack and Vagrant for Drupal Development | May 20, 2013 |
| Making Linux and Android Get Along (It's Not as Hard as It Sounds) | May 16, 2013 |
| Drupal Is a Framework: Why Everyone Needs to Understand This | May 15, 2013 |
| Home, My Backup Data Center | May 13, 2013 |
- I once had a better way I
5 hours 20 min ago - Not only you I too assumed
5 hours 38 min ago - another very interesting
7 hours 31 min ago - Reply to comment | Linux Journal
9 hours 24 min ago - Reply to comment | Linux Journal
16 hours 18 min ago - Reply to comment | Linux Journal
16 hours 34 min ago - Favorite (and easily brute-forced) pw's
18 hours 26 min ago - Have you tried Boxen? It's a
1 day 17 min ago - seo services in india
1 day 4 hours ago - For KDE install kio-mtp
1 day 4 hours ago
Enter to Win an Adafruit Pi Cobbler Breakout Kit for Raspberry Pi

It's Raspberry Pi month at Linux Journal. Each week in May, Adafruit will be giving away a Pi-related prize to a lucky, randomly drawn LJ reader. Winners will be announced weekly.
Fill out the fields below to enter to win this week's prize-- a Pi Cobbler Breakout Kit for Raspberry Pi.
Congratulations to our winners so far:
- 5-8-13, Pi Starter Pack: Jack Davis
- 5-15-13, Pi Model B 512MB RAM: Patrick Dunn
- 5-21-13, Prototyping Pi Plate Kit: Philip Kirby
- Next winner announced on 5-27-13!
Featured Jobs
| Linux Systems Administrator | Houston and Austin, Texas | Host Gator |
| Senior Perl Developer | Austin, Texas | Host Gator |
| Technical Support Rep | Houston and Austin, Texas | Host Gator |
| UX Designer | Austin, Texas | Host Gator |
| Web & UI Developer (JavaScript & j Query) | Austin, Texas | Host Gator |
Free Webinar: Hadoop
How to Build an Optimal Hadoop Cluster to Store and Maintain Unlimited Amounts of Data Using Microservers
Realizing the promise of Apache® Hadoop® requires the effective deployment of compute, memory, storage and networking to achieve optimal results. With its flexibility and multitude of options, it is easy to over or under provision the server infrastructure, resulting in poor performance and high TCO. Join us for an in depth, technical discussion with industry experts from leading Hadoop and server companies who will provide insights into the key considerations for designing and deploying an optimal Hadoop cluster.
Some of key questions to be discussed are:
- What is the “typical” Hadoop cluster and what should be installed on the different machine types?
- Why should you consider the typical workload patterns when making your hardware decisions?
- Are all microservers created equal for Hadoop deployments?
- How do I plan for expansion if I require more compute, memory, storage or networking?







Comments
Virtualization?
I think the details of a PPC port of Linux KVM are a bit sketchy, but can you imagine buying one of these as a VM host? That would be interesting. I know historically PPC processors are better at math and floating point operations.
-- FLR or flrichar is a superfan of Linux Journal, and goofs around in the LJ IRC Channel