(Participation in this project is FREE. It makes use of unallocated resources within your PC and causes no detectable degradation of performance.)

One in four of us will at some time suffer from cancer. The high mortality rate, the suffering experienced by patients from the side effects of existing treatments, and the high costs of treatment all contribute to making cancer a priority for drug research.

By joining the United Devices network, you can have a direct benefit to cancer research. The UD software sits in the background on your PC and utilises any spare capacity to analyse data for cancer research projects. (It can also be set just to run when your computer is idle if you are concerned about your PC performance being compromised, but, in reality, there is very little overhaead and even running live you should see no degradation.)

Each individual computer analyzes a few molecules and then sends the results back over the Internet for further research. This project is anticipated to be the largest computational chemistry project ever undertaken and represents a genuine hope to find a better way to fight cancer. (You can also choose to work on anthrax or smallpox.)




------------------------------------------------------------------------------

The link to the site is http://www.grid.org/projects/cancer/

Downloading the agent takes only a few minutes and once it is done it will function quite happily on its own. You can make adjustments to how and when it runs very easily. There is a very comprehensive FAQ section to help.

I have set up a MoneyAM BBers team so that we can pool our work and statistics. Why not join in and see how much of a difference MoneyAM can make?

The link to our team is MoneyAM Users Team - JOIN NOW
-----------------------------------------------------------------------------

Join us today and make a difference while you make money. One day you might be grateful to take advantage of the developments this research achieves.

Don't just make a donation, make a difference!

-----------------------------------------------------------------------------

PS Thanks to MegaBucks who suggested the idea of starting here.

skg - yes, it was. Now on a cancer project, so the system must be up and running again.

Down again

Posted: Tue Nov 29, 2005 11:10 pm Post subject: Grid.org outage

--------------------------------------------------------------------------------

Dear Members,

We have incurred a major outage at our new facility due to a bad circuit breaker. This caused everything including the forums to be down for a while. We are in the process of recovering now. All of the servers are still not up which is why everyone is getting the backing off message. This will be corrected as soon as possible.

This outage is not related to the old hardware, but an issue related to the facility move. We did not anticipate this outage, but with a facility move of this magnitude, some temporary glitches are not surprising. Please bear with us while we work through these issues and again thank you for your contribution.
_________________
Robby Brewer
Senior Support Engineer
United Devices

This now a regular event. I normally have to re-register when it comes back on.
According to them I have about 18 machines registered but I only have 4.
Is there anywhere else that I can let use my spare capacity to achive the same results like helping someone to find cures.
Graeme.

Graeme
I think that we all suffer from the same frustration. A least this latest disruption did not create another set of phantom machines as previous ones have. For me there are pros & cons to moving to another project. The current UD Cancer Research Project meets most of my criteria for contributing:
1. I am directly participating in a cancer drug discovery project
2. It is for a UK academic organisation
3. Being in a team creates an active participation through competition

After over a year of contributing CPU time to the project we should now have a comprehensive list of cons, mainly centred on the not infrequent site outage problems.

These really play on the the conflicting problem of running machines 24/7 with the subsequent waste of energy. I personally find the breaks useful as I use the down time to perform additional maintenance and the turn them off and give them a rest from running at 100% load.

Another alternative is to see if it is possible for one or more of the team to keep a store of work units that could be distributed if the UD site is down.

It would be shame if individuals or the team felt that it was time to move to another project but I have found the following site, the Center for Distributed Computing has a list of similar projects.
skg

skg
Thanks for your coments. You sumed it up with the word "frustration", I find it hard to understand why more people haven't signed up to this when it appears that if we all made a little effort / contribution by signing up, then huge problems like a cure for cancer could be found. (from little acorns great Oaks do grow). But then I become frustrated when I think such a great opportunity could be lost because of what appears to be poor performance by the people running it. I will hang fire for the moment to see if things improve, then if they don't then I will see if my tinyy contibution could be beter used by someone else to achive a similar goal.
Lets all keep crunching and trying to make a difference.
Graeme.

Not sure why you're having to reregister Graeme, the only time I ever had to was when they lost their whole database. I just ignore it and it reconnects by itself eventually.

I have done a bit of research, there is a program called UD Monitor that will download and manage multiple work units. I will install it over the weekend and comment back on it's usefulness.
skg

Kayak, Thanks.I have a total of four machines but now show 18. I can understand +2 because of new hard disks and +4 if every one had to re register, that makes 10 but I don't know where the other 8 have come from or why. There are only four active.
The other problem ( I now just ignore) is in device manager I cant set connection type and they are all showing not known because anything else I try I get duplicate device message error.

skg.
I like the sound of that, at least I won't be wasteing power leaving them on all the time. I will look for your update after you have tryed it. Cheers. Graeme.

Graeme and skg, hope you two keep up the good work, the team will look distinctly weak if you quit! I thought I lost points in the summer downtime, so recently when I couldnt check progress I logged off f.t.b. Ive just got the one machine (perry97) are you using a special number cruncher? If this is part financed by a British university maybe theres a shortage of funds to meet demand until the numbers click bringing the breakthro we all hope for. I am down as having 7 machines !!

Graeme
You can rename the old duplicate devices.
e.g. I renamed the two old duplicate devices skgpc_1 to skgpc_1z and skgpc_1y
You can then set the connection type for your current device

skg

I have downloaded the latest version of UD Monitor. I am testing it out to see how useful it is managing multiple work units on multiple PC's.

skg

Hi skg,
any update on how you are getting on with the latest version (post 101).
I am back to having problems conecting, downloading and displaying stats. one machine hasn't been able to conect for weeks. Graeme.

I am using the latest version of UD Monitor. You can create "Slots" which are complete copies of UD with an associated work unit and require a new device identifier. I have not tried it yet but you could copy the folder containing a "Slot" to another machine. The reason I haven't tried it yet is UD Monitor has "lost" the link to the Slots that I've created. It doesn't seem to be having a negative impact in my current set up, i.e. continuing to process the work units but I lost control of the additional WU's. I see this as teething probs and will sort it out and report when I return home at the end of the week. This is definitely the way to keep a supply of WU's to fill in time between site outages.

Graeme email me if you want a more detailed discussion.

skg

ttt.

An update on running UD Monitor
UD Monitor is a complementary program to manage multiple work units.
It allows a user to download multiple work units and work on them in order.
This is done by UD Monitor by creating "Slots" to form a cache of work units. Each of these Slots needs to be registered as a different device. This is inconvenient but it is how United Devices keeps track of the work units it has distributed.
Each of the Slots is a complete copy of the United Devices folder including a single work unit.
UD Monitor works by monitoring the files in the United Devices folder and when a work unit has been completed, the results uploaded and a new work unit downloaded, it stops UD.exe and copies all the files in the United Devices folder to the current Slot (a numbered sub-folder in the UD Monitor Cache folder). UD Monitor then copies all the files from the next Slot folder (in date order) to the United Devices folder. UD.exe is then restarted and processing of the next work unit commences.
There is a little configuration allowed in UD Monitor, particularly there is no choice as to the location of the cache. This means that local distribution of the work units would require additional steps. (you can execute files on completion of a work unit so you could create a batch file to copy a local slot to a central cache)
As the main reason for exploring the use of UD Monitor was to cover periods when communication with the UD server was not possible I think that running UD Monitor on one machine to hold a central cache. Work units have to be completed within 450hrs (18 and a bit days) so you could store the maximum number of cached work units on your fastest PC but the more slots and work units you have the more devices you need to register.
In the event of a loss of communication with the UD server you can store a completed work unit from a PC without UD Monitor to a new slot in the central cache and copy a new work unit back, marking the slot inactive to stop it being processed twice.
I will continue to experiment to see how easy or not it is to automatically distribute work units locally.
Please comment if this is too long and complex or feel free to ask questions.

skg

skg.
Thanks for the update above. I have spent some time over the last week trying to get my head around this. I have now downloaded it but it was only tonight I finally got my head around "the slots bit, I didn't right click in the cache window". Mission tomorrow to set up the slots on my laptop the see how it runs for a few days. If all seems ok I think the easyiest way for me will be to put the monitor on all four machines and treat each one as stand alone.

I finally managed last week to get them all working again last week after the last lot of problems so really hope I don't end up in a mess again.

I think in the near future we may experience quite a lot of downtime / non-connection as they are going to install new harware and then new software.

Thanks for your posts and help. Graeme.

Have now created 6 slots / caches on the laptop and all seems ok. fingers crossed for the next few days as I monitor how long it takes to work through them. Graeme.

In terms of points generated, our little Team, with 4,075,988 is now in 875th position worldwide... Quite a performance!

G.

This is due to the teams contribution of 22 years 7 months cpu time to the cancer research project.
Anyone interested can join the MoneyAM BBers team, just download and install the UD software and then go to MoneyAM BBers team page to join.
skg

any one else having problems conecting the last few days.
I can't even get to the message boards. Graeme.