My First Experience with TN-Grid
Hi everyone,
This morning I decided to join for the first time the TN-Grid project, where I discovered that the Gridcoin Team is the major contributor of the project (five times more points compared to the second entry in the top).
TN-Grid is a research project in genetics, hosted by the University of Trento, in Italy.
The project is using an application (gene@home) that is building a network of casual relationships between every two gens. The algorithm finds a gene network and studies its causal relationships, in order to estimate if a list of new genes can have a causal relationship with already known genes.
Two applications are currently running on the TN-Grid project: Vitis vinifera (OneGenE), studying the biology of grapevine, and Pseudomonas Aeruginosa (OneGenE), studying the genetics of the multidrug resistant bacterium.
Since morning, I already received 27 work units. Out of those, ten are already completed, while two got also validated. What I can see so far, each work units takes approximately 3.5 hours of CPU time (on a i5-3350P CPU @ 3.10GHz, running Ubuntu), and is giving out 165 credit points.
I'm just curious, is anybody else experimenting, these days, with TN-Grid? If yes, what is your experience?
TN Grid is my main project. It's really good for ARM devices. I'm hoping to crack the top 100 in total credit soon. Currently ~150th.
TN-Grid is one of my favourite BOINC projects. I love the genetics aspects of it. I'm in the grcpool and it gets me the most MAG after Milkyway@Home.
Honestly I got into TN Grid after I read a different post showing it to give very high RAC for CPUs, but after reading into it I really love the project intent overall and have switched almost exclusively to it on all of my hosts. In my experience though I have needed to keep a second project attached as I have ran out of WUs a couple times when exclusive on TN...
I tried TN-Grid aswell, it has quite high payouts from how it seemed to me, i tried it with an old cpu and the generated Magnitude was quite good for old devices, I think it's also good for Starters.
I read that this application is giving very interesting results on ARM. Today will try it on a old Android mobile phone...
I have done SETI@Home and Folding@Home on my old PS3 but haven't tried this one. Looks like I may have a new app install in my future.
Seti@home and Folding@home are giving very good results on GPUs, while TN-Grid is CPU exclusively. You should have this in mind if you decide to try it.
I have been crunching for TN-Grid for a couple of months now.
When I joined work units were hard to get because the tasks could be completed faster than they were being generated.
Since research on Pseudomonas Aeruginosa began work supply has been steady with my i7 PCs taking 3-4 hours per work unit.
I receive work for both the AVX and SSE2 optimized applications which seem to complete work units at the same rate as eachother.