The Musings of Chris Samuel

This coming Saturday, 21st August 2010 will be my first opportunity to participate in Australian democracy. My citizenship came through a few months after the last election, had I’d been able to vote then I’d have cast my vote for Labour and against John Howard.

However, with the Australian Labour Party (ALP) lurching to the right on a number of issues such as immigration, continuing the failed intervention in the Northern Territory, failing to legitimise same-gender marriage, and their crazy idea of mandatory Internet censorship combined with a new do-nothing strategy on climate change (“let’s hold a citizens assembly to tell us what to do, just like we did in 2008!”) means my conscience does not permit me to give them my first preference. They at least have some vision with the NBN, but that’s about it.

As for the Coalition, well they’re just laughable. A leader who doesn’t understand science or technology, policies that promise to deliver half the current speeds of ADSL2+, obscene exaggeration and fear-mongering about refugees coming in by boat (here’s some much needed facts on the matter), wanting to make bible classes compulsory in schools (I suspect aimed at the even more right wing Family First to whom they are directing preferences) and even worse policies on climate change and greenhouse gases. Even more FAIL than Labour.

So, I’m voting Green because:

• They want to enshrine basic human rights in law (Australia is the only western democracy without legal protection of freedom of speech)
• They’re against the mandatory Internet censorship scheme
• They take the science of climate change seriously, and the challenges it poses
• They believe that people who love each other should be able to get married, irrespective of orientation
• They wish to treat refugees as people, not some mythical threat
• They understand free, open source software and use it themselves

Most importantly I’m voting Green because THEY WANT YOU TO THINK! Not just about their policies, or other parties policies, but to think about how you direct your preferences. Sure they have preference deals, but what most impressed me was when they were announced Bob Brown said:

I don’t like back room preference negotiations with other parties. In fact I’m sick of it. And I think that we should be very well aware here that voters can get misled into believing that they should put their preferences where the Labor party or the Liberal party or the Nation party or the Greens or somebody else says. No that’s not true. People have a right to put, and I think an obligation to think about it, and put their preferences where they want to. That’s what’s important.

Watch the video on that ABC news article to hear that, it’s sadly not in the text of the report. They also have the best election advert that never was – The Gruen Transfer has been getting two advertising agencies a week to do an advert each for a political party and this one won the week they did The Greens.

Now I’m not under any illusions that they’ll form the next government, but voting for them will send a signal that I’m not happy with either of the major parties, and they should (hopefully) get the balance of power in the Senate.

Last week I was invited to take some publicity shots for the Software Freedom Day Melbourne crew at the State Library of Victoria Experimedia centre. Asides from occasional complaints from my camera (the infamous Nikon ERR CHA happened 3 times) I managed to get about 200 shots which I’ve whittled down to 26 of the best and put them up as a set on Flickr.

All done with free software (well, asides from the firmware in the Nikon D90) – Digikam rocks!

The Melbourne Free Software Interest Group (a group of Melbourne computer folks with an interest in software freedom) have put together an open letter to Senator the Hon Kim Carr, the Minister for Innovation, to request that software be excluded from patenting as part of the Australian governments review of patents in general.

We are currently collecting signatures to the letter and if you are in Australia and of a like mind we would really appreciate it if you would contribute your signature too! Just click on the link, read the letter and the form to sign it is at the bottom of the page. Please also pass this on to others you know who may be interested.

The most excellent Joe Landman has a blog post looking at the performance of an anonymous SSD with a Sandforce SF-1200 controller chipset on it, and comes up with some very interesting benchmark results. But first he comes up with a nice way of quantifying the distance between benchmark results and real life:

I need to understand how close the marketing numbers are to the actual numbers. We need to establish a ratio for this. Call this the Benchmark Significance Ratio, or BS Ratio for short. Define BS Ratio as:

BS Ratio = (what they claim) / (what you measure)

A BS Ratio close to 1 is good. A BS Ratio much greater than 1 is bad. Of course, a BS Ratio much less than 1 is either an indicator of a failed test, or an accidentally released product.

What Joe finds is that the performance of the SSD, in terms of basic things like read/write speed, depends on what you write to it. If you write lots of zeros you find the performance is almost 4 times as much as if you write random data to it. Now as Joe rightly points out, this smacks of compression somewhere in the path between the program and the disk which means that most of the benchmarks you see/do (unless they/you, as Joe does, take care to use random data) will be pretty much meaningless unless you plan to just store zeros. Mind you if you do plan to just store zeros then I suggest just using /dev/null for writing to and /dev/zero to read from – they will give you much better speed and far better capacity for free!

What intrigues me though is not so much the speed difference but what that means for the capacity of the device – does it claim to be a X GB device but actually store Y GB (where Y > X dependent on compressibility of data) or does it enforce the amount that it can store to the quoted capacity ? Even worse, does the stated X GB capacity depend on your data being compressible and more random data results in less space ? I think the last can be ruled out because I can’t see a way how you would fake that failure to the SATA layers unless you returned failed writes which could cause chaos, especially in a RAID environment. I also suspect that it must enforce a fixed limit as I presume they must fake some characteristics of a spinning disk (heads, etc) for compatibility.

Which is a real shame because you’ve actually got a storage device that could (given the right sort of data) store far more than its stated capacity, if only you could address it in a non-spinning disk like manner. This echo’s other issues with SSD FTL’s like bad wear levelling implementations, etc, which would go away if we had an open interface into the device exposing the internals, that way you could (with this controller) get extra storage for free and potentially even a better wear levelling system into the bargain.

I guess even if that were available I don’t know if filesystems could cope with that (yet), but I wouldn’t mind betting they’d be up for it!

Update 2: Borders Australia have now released an SD card image for upgrading under Linux. Well done Borders AU!
(more…)

For folks who’ve not yet noticed, the long foreshadowed dropping of Xen from RHEL6 for anything other than as a guest OS is still on:

Red Hat Enterprise Linux 6 Beta is supported as a Xen guest for the x86 and x86_64 architectures. Additionally, Red Hat Enterprise Linux 6 Beta can be used as both a Xen paravirtualized (PV) guest or as a Fully virtualized (FV) guest with PV drivers. Due to paravirtualized operations (pv-ops) being included in the kernel, the same kernel can be used for either mode of operation as well as for bare metal. There is no support for using Red Hat Enterprise Linux 6 Beta as a Xen host.

If you are using Xen with RHEL as the base for your virtualisation you’ll be wanting to look for alternatives..

If you’re running KDE 4.5.0 beta 2 from the Kubuntu PPA and are wondering why your widgets disappear every time you login, it’s because beta 2 has a bug where it gets really confused about which activity you are using – use meta-tab (windows-tab) to switch between them and you should eventually find what you’re looking for..

Also if you’re wondering why the “Printer Settings” option isn’t working any more and giving a strange error message, it’s because it’s now looking for a Python module which isn’t packaged in Ubuntu (kubuntu-ppa bug 591980), it’s complaining that:

systemsettings(22750)/python (plugin): Failed to import module
 Traceback (most recent call last):
   File "/usr/share/kde4/apps/system-config-printer-kde/system-config-printer-kde.py", line 64, in 
     import cupsutils.ppds
 ImportError: No module named cupsutils.ppds
 systemsettings(22750)/python (plugin): Failed to import module

In Debian the module it’s looking for is in the “python-cupsutils” package, but that’s not in Ubuntu.

One of the things that us HPC folks tend to get hot under the collar about is hardware locality, basically making sure that your memory accesses are as fast as possible by optimising where on the system you’re getting memory from and making sure your process doesn’t get moved further away. Just binding your processes to the cores they are on can make for a significant speed up so it’s well worth doing. If you’ve just got a single socket, or a pre-Nehalem Intel x86 system then your path to RAM has been pretty much identical wherever you are so the only benefits are from not moving away from your CPU cache lines but on AMD Opteron, Nehalem, Itanic, Alpha, etc you really should care a lot more about locality for best performance.

The open source Torque queuing system (which I help out with) does some of this already, if you compile it with –enable-cpuset and have the /dev/cpuset virtual filesystem mounted then before it starts a job on a node it will create a cpuset for that (based on what cores have been allocated on the node) and then put the HPC processes into that cpuset. If you’re using Open-MPI 1.4.x and have the environment variable OMPI_MCA_orte_process_binding set to core then each of the MPI ranks will bind itself to one of the cores within that cpuset.

All good ? Well not quite as Torque is reliant on /dev/cpuset being there and being able to parse the contents of it and Open-MPI 1.4.x uses the Portable Linux Process Affinity (PLPA) library which, as its name suggests, is only for Linux. So the good Open-MPI people looked at their PLPA library and decided it needed extending and teamed up with the INRIA libtopology team who were working on how you discover the topology of various architectures and decided to merge the two projects together under the banner of the Portable Hardware Locality (hwloc) library.

The Portable Hardware Locality (hwloc) software package provides a portable abstraction (across OS, versions, architectures, …) of the hierarchical topology of modern architectures, including NUMA memory nodes, sockets, shared caches, cores and simultaneous multithreading. It also gathers various system attributes such as cache and memory information. It primarily aims at helping applications with gathering information about modern computing hardware so as to exploit it accordingly and efficiently.

The portable bit of the name comes from the fact that it works on Linux, Solaris, AIX, Darwin, FreeBSD, Tru64, HP-UX and Windows (though with limitations on some architectures – e.g. Windows – which don’t expose all the info it needs) and can extended for other OS’s if people feel they need to scratch that itch (OpenVMS anyone?). This release is also embeddable into projects (such as Open-MPI 1.5) and I have an interest in Torque picking it up to improve and extend its cpuset support.

Well another battle in the long war with SCO is over, a US court jury trial has confirmed that Novell do indeed own the UNIX copyrights and SCO do not. Of course SCO could appeal, but this would be the second time they’ve lost on this point.

For those computer history buffs it is sad to learn that the Linux kernels DECnet code is going to be orphaned in 2.6.33, the git commit by Christine Caulfield says:

Due to lack of time, space, motivation, hardware and probably expertise, I have reluctantly decided to orphan the DECnet code in the kernel.

Judging by the deafening silence on the linux-decnet mailing list I suspect it’s either not being used anyway, or the few people that are using it are happy with their older kernels.

To be honest I’m surprised it’s lasted this long, the last time I used DECnet in anger was around 1997 I think..