Even though the GPU task is running at less than 1%, is there any significant dropoff in CPU efficiency in terms of increased time to finish the CPU task? I would think we are talking maybe 20% slower(?)
The numbers are a little inconsistent and people are reporting a loss of 3-9% decrease in throughput ...
I looked at a total of 14 tasks 6 of 6.61 and 8 of 6.62 and came up with the 9% figure ... personally I am happy with it because I regained 21% CPU that had been lost under the 6.61 where the poll loop was consuming so much CPU.
To my mind the loss on the one side is more than made up with the gain on the other... so far most other people agree ...
I suggested that the project consider an option where people could select their own preference where those that are doing ONLY GPU Grid could select the option similar to 6.61 where they do not care what the CPU load is ... and the "nice" option for those like me that prefer a lighter load on the CPU side of GPU tasks so that we can get the most out of the system as a whole.
If I need more throughput I just need to add more GPU cards ... :)
RE: Even though the GPU
)
The numbers are a little inconsistent and people are reporting a loss of 3-9% decrease in throughput ...
I looked at a total of 14 tasks 6 of 6.61 and 8 of 6.62 and came up with the 9% figure ... personally I am happy with it because I regained 21% CPU that had been lost under the 6.61 where the poll loop was consuming so much CPU.
To my mind the loss on the one side is more than made up with the gain on the other... so far most other people agree ...
I suggested that the project consider an option where people could select their own preference where those that are doing ONLY GPU Grid could select the option similar to 6.61 where they do not care what the CPU load is ... and the "nice" option for those like me that prefer a lighter load on the CPU side of GPU tasks so that we can get the most out of the system as a whole.
If I need more throughput I just need to add more GPU cards ... :)