bv2.co.uk

Last night I received an email asking how to install the BV2 Thermal Monitor application. The simple answer is:  you don’t.   I can see that it may not be clear from the original post. As the application is entirely self contained and does not depend on any C runtime you only need to:

extract the zip file to a directory of your choosing

double click the .exe file

You do need to have the NVidia drivers installed. You can make desktop shortcuts / start menu shortcuts to it in the usual manner. Please post any more queries to the forums.

The forums are back online :)  I switched from BBpress to phpBB, not that there is anything wrong with bbpress but I am more familiar with phpBB and the various options etc in the setups. phpBB is a bit overkill for such a little site but all the features and options are rather fun.

The three users who registered on the old forums will need to re-register but there were no posts apart from mine - so no loss there

I’ve had the odd email go missing from the forums but I think I’ve sorted out the mail server properly now. Please let me know if your registration mail goes missing.

For some reason most vistors to the site tend to email me rather than leave a comment. I really don’t mind and try and answer them as quickly as possible. If you don’t get a reply within a day the chances are my spam filter has eaten them…

In order to promote more open discussions about the topics on this site I have installed some forum software from bbpress (the makers of wordpress). You will find the forums here or click on the link on the top right hand of every page. BBpress integrates into wordpress so your blog user accounts should work in the forums too.

I am moving the employment / contract pages to the forum so people can post things themselves. The forums are protected by spam filters etc but if something does get through and I don’t notice - please let me know.  Although at the moment there is very little chance of me not noticing as there are almost no topics!

Finally v0.1 of the BV2 GPU Thermal Monitor is ready for download. It is available in the downloads section of the site. The source code is included! Please note that I only partially translated the nvapi.h file to an .inc file.

Please note that it is licensed differently from other content here and is under the Creative Commons “Attribution-Share Alike 2.0 UK: England & Wales” License. Basically this means you are free to copy, use and modify the code/application in any way you like as long as: you give the original author credit and if you alter, transform, or build upon this work, you may distribute the resulting work only under a licence identical to this one.

By downloading you agree to be bound by the license.

Ok legalities out the way

BV2 Thermal Monitor:

BV2 Thermal Monitor v0.1 Screenshot

Currently it is version 0.1 and does basic temperature monitoring of up to two NVidia GPU’s.

It updates approximately every 500ms.

It can be minimized to the system tray where hovering the mouse over the icon will display a tooltip with the temperatures detected. (more…)

I was hoping to release the GPU temperature monitor to the downloads section sometime during this last bank holiday weekend. I had also planned to sort out my somewhat ailing / overheating computer and perform some upgrades. Unfortunately the repair / upgrades took almost 2 full days. My pc is now running a lot cooler and a bit quieter and I am now something of an expert with heat spreader / heat sink cleaning and thermal grease application.  By the way artic silver is really good! Even before the 200 hour break in period mentioned on their site I’m already seeing more than 10 degrees lower temperature on the CPU.

As to the upgrades:  more posts on this later But it did include me purchasing Windows XP Professional 64bit.

The longer than expected pc maintenance time has impacted the GPU Thermal Monitor application and it won’t be ready for a few more days. A bit of good news about it though:  it works perfectly on Windows XP 64bit without any modifications. Don’t you just love asm :)  Some of my other C/C++ applications were not quite so happy on the new OS.

Keep watching this space for release date

As part of our “honeymoon” Catriona and I did a series of day trips around the South and Midlands of our not-so-little island. We are both members of the National Trust but don’t often have time to visit all the locations on offer. We saw some lovely homes and parks and with Catriona having studied history she made a very capable tour-guide

One place that is worth a special mention is Bletchley Park. Sadly, it isn’t part of the National Trust and receives no Government Funding (well done to Milton Keynes Council for giving £600,000 for urgent restoration).

(more…)

(update 27/5/09 - I’ve now released v0.1 - see this post )

It’s been a while since my last post as I’ve been rather busy with work and some other projects - so time for a quick update. 

While working on some long running kernels I wanted to keep track of the GPU’s temperature and ran the monitor app that came on their installation disks. The problem with the monitoring apps that I have is that they are quite large and take up a lot of screen space and for some unknown reason they crash / stop working when viewed over VNC (or similar) remote control app.  Now as I mostly use my CUDA machines over the network this is not a good situation.

So decided to write my own :)  Here is a screenshot of the initial version. It still needs config and some cleaning up (can you spot the glitch by the minimize button?) and possibly some remote reporting over the network functionality. Currently it can display the temperature from two GPU’s and updates every 500ms. When minimized it sits in the system tray and updates its tooltip with the reported temperatures. I will be releasing it to the downloads section of the website sometime over the weekend. The current .exe is 146k (mostly the skin) and doesn’t require any installer.

GPU Thermal Monitoring Application

Nothing wrong with a bit of self promotion sometimes

In my post a few days ago I mentioned that some of the numbers reported by the Visual Profiler needed further investigation.

In particular the device to device memory bandwith reported by the profiler differed from the value reported by the bandwith test sample. This was easily tracked down:  according to the documentation, the profiler divides the bytes read/written by 10^9 whereas the bandwith test sample divides by ( time in milliseconds * 2^20).    So: 1000*2^20/10^9 = 1.048576    i.e. the bandwith sample reports a lower number.

(more…)

I managed to get some time last night to convert my LBM implementation to CUDA. Its far from optimal at the moment. Here is a screen shot showing the lid-driven-cavity with two obstacles, one horizontal and one vertical in the box. The visualization is from the middle plane. The obstacles are not shaded but are easily seen in the picture as an area of black. This image was taken after 16000 timeslices and some nice stable vortices have developed.

D3Q19 - lid driven cavity with obstacles

Once again the importance of making a gold kernel cannot be understated as I had quite a number of bugs in my initial CUDA implementation. I use a “multi-tap” type approach when debugging where the kernels write intermediate results to device memory as they go along. This can be easily compared to data coming from the various stages of the gold kernel and makes it much easier to identify the source of the error. Keep in mind the CUDA floats will never be the same as the CPU’s floats (CPU uses 80 bits to compute intermediate results).

According to the CUDA programming guide the memory coalescing rules have been relaxed in devices with compute capability 1.2 or greater.  Chapter 5 has a subsection entitled “Coalescing on Devices with Compute Capability 1.2 and Higher” which gives more information.

As I now have a C1060 which has compute capability of 1.3 I thought I’d run my old coalescing tests on it to see how it has improved.  I modified the launch configuration to 32760 blocks in order to maximize utilization of all the Multiprocessors and increased the thread count to 256. These changes cause the kernels to report 100% utilization in profiler.  I expected the uncoalesced part of the old tests to be quite kind to the Tesla as they fall within the parameters given in the programming guide and indeed the memory transfer rates were similar to a pure device to device copy. I then modified the uncoalesced kernels to avoid the auto coalescing of memory accesses in a half-warp.

(more…)