Message boards :
News :
Moving to fixed credit
Message board moderation
Author | Message |
---|---|
Send message Joined: 8 Jul 11 Posts: 1344 Credit: 533,243,450 RAC: 547,572 |
I have replaced CreditNew (or CreditScrew as some like to call it) with a fixed credit system. The fixed value is based on the average runtime from many samples from a single machine in a well controlled environment (one of my own computers). Obviously, the fixed value will change with each batch, since each batch has a different average runtime. This should be a pretty fair system, especially after a host has processed enough WUs for the average to kick in. The main downside is the potential for cherry pickers - hopefully there won't be too many of them. |
Send message Joined: 26 Jun 13 Posts: 11 Credit: 8,735,592 RAC: 0 |
Given that workunits for this project are either extremely short or steadily long, I won't be surprised if cherry pickers rush in because it's too easy to sort out those long tasks. Are there any motivation behind this? |
Send message Joined: 19 Aug 11 Posts: 31 Credit: 73,721,046 RAC: 2,712 |
|
Send message Joined: 8 Jul 11 Posts: 1344 Credit: 533,243,450 RAC: 547,572 |
Given that workunits for this project are either extremely short or steadily long, I won't be surprised if cherry pickers rush in because it's too easy to sort out those long tasks. Are there any motivation behind this? I think you are remembering the old days. The current batch has a mean of 1.5 hours with a variance of about .5 hours (3.6GHz single threaded). It would be quite tedious to sit there and cherry pick WUs just to increase your credits by a small margin. The server will also cut off a host whose error rate is too high, and I could lower this threshold if I needed to. The motivation is to address complaints from users regarding poor credit. CreditNew treats people unfairly, especially in the beginning as the stats settle. There's also an issue with people cheating - in a nutshell, benchmarks can be spoofed to make a CPU look much better than it really is, and this results in higher credits. With fixed credit, you can tweak your benchmarks all day long and it will have no effect. |
Send message Joined: 27 Jul 13 Posts: 1 Credit: 904,692 RAC: 0 |
This is fantastic news, CreditNew has never worked properly and it's so much nicer to know that you're actually getting credit proportional to the work that you're completing. |
Send message Joined: 18 May 18 Posts: 1 Credit: 3,779,308 RAC: 0 |
Great news. CreditScrew has lived up to its name and should eliminated. |
Send message Joined: 15 Jun 15 Posts: 2 Credit: 1,194,710 RAC: 0 |
Very very good news! Thanks. |
Send message Joined: 26 Jun 13 Posts: 11 Credit: 8,735,592 RAC: 0 |
Ah thanks for the explanation. I should come back to have a try after finishing the current batch of work on my hand. In fact, I wondered why there aren't short tasks anymore some time ago, but that explains it now. |