Posts by Eric Driver

1) Message boards : Number crunching : NF is this week's Formula BOINC's Sprint (Message 2188)
Posted 1 day ago by Profile Eric Driver
Post:
Now, wait until you get GPU code running.


Funny you should mention that... I just got a prototype working on my GTX 1050. It's about 20 times faster than the CPU version. Probably still a few weeks out before it's production ready though.


Awesome! I assume that if the GPU is 20 times faster than CPU, then you will get your result quicker too, assuming the calculation method is the same with both CPU and GPU(?). Please keep us posted. It is exciting to know what's coming soon and I'm sure some crunchers here are ready to fire up their GPUs when the task are available.


Yep, results should be that much quicker. It works with the same exact WU inputs. On my test case, the CPU time was 13000 sec, the GPU time was 500 sec.

I will open a new thread in the next several days to post more info, and maybe put a beta version out there.
2) Message boards : Number crunching : NF is this week's Formula BOINC's Sprint (Message 2185)
Posted 2 days ago by Profile Eric Driver
Post:
The free-dc stats page doesn't have all of the top users on it, which could explain the discrepancy. Compare to this:
https://numberfields.asu.edu/NumberFields/top_users.php
I wonder if free-dc stats only uses data from those who consented to stats export.
3) Message boards : Number crunching : NF is this week's Formula BOINC's Sprint (Message 2183)
Posted 2 days ago by Profile Eric Driver
Post:
Now, wait until you get GPU code running.


Funny you should mention that... I just got a prototype working on my GTX 1050. It's about 20 times faster than the CPU version. Probably still a few weeks out before it's production ready though.
4) Message boards : Number crunching : NF is this week's Formula BOINC's Sprint (Message 2181)
Posted 2 days ago by Profile Eric Driver
Post:
You're welcome. And thanks to all the participants!
5) Message boards : Number crunching : NF is this week's Formula BOINC's Sprint (Message 2178)
Posted 3 days ago by Profile Eric Driver
Post:
I have some information regarding server load. I think a good measure of that is how many WUs are processed per hour. Here is an average, broken down by day:
Before Sprint (3-13): 3571 WUs/hr
Thurs (3-14): 9329 WUs/hr
Fri (3-15): 5042 WUs/hr
Sat (3-16): 4311 WUs/hr
Sun (3-17): 6593 WUs/hr

On Friday, the current batch that had been running (14x11) ran dry and processing started on the next batch (14x12). The WUs in 14x12 average twice the run time as 14x11, so to compare apples to apples, I did a 2nd calculation where I double the WU rate for the 14x12 cases before adding to the 14x11 rate. This gives the following WU rates (this is what the numbers would have looked like if both batches had the same run times):
Thurs (3-14): 9329 WUs/hr
Fri (3-15): 6871 WUs/hr
Sat (3-16): 7930 WUs/hr
Sun (3-17): 10692 WUs/hr

I think the key point here is that the worst server load is at the start and end of the sprint. I'm not an expert on the strategies behind these competitions, but I assume it's due to bunkering up front and then holding back results until just before the finish. Anyways, the other point of interest is that the worst server load was about 3 times our typical load.
6) Message boards : Number crunching : BSOD's in my last batches (Message 2177)
Posted 4 days ago by Profile Eric Driver
Post:
Things look more normal now from my end. At least you have tasks and NumberFields is on your list of projects now. Let me know if you still get the BSOD. I haven't had any other BSOD complaints.
7) Message boards : Number crunching : NF is this week's Formula BOINC's Sprint (Message 2175)
Posted 4 days ago by Profile Eric Driver
Post:
Found via the rechenkraft.net forum:
The beginning of this sprint brought NumberFields@home's second and third highest credits/day in boincstats' recorded history, eclipsed only by the first day of Pentathlon 2018.


Very interesting! Thanks for the link.
8) Message boards : News : Septic Results Paper Accepted for Publication (Message 2173)
Posted 5 days ago by Profile Eric Driver
Post:
Very nice, indeed. Please post a link to the paper as soon as it becomes accessible.

Michael.


Will do.
9) Message boards : Number crunching : NF is this week's Formula BOINC's Sprint (Message 2169)
Posted 7 days ago by Profile Eric Driver
Post:
I saw the tasks in progress growing fast, so I dumped a bunch more WUs into the queue. I'd rather have too many out there than not enough.
10) Message boards : Number crunching : NF is this week's Formula BOINC's Sprint (Message 2167)
Posted 7 days ago by Profile Eric Driver
Post:
Hi Eric,

Out of curiosity, would you be able to publish an estimate of the server load and completion of WU increase before and after the FB sprint event? There are some projects where the server and/or WU generation can't cope up during the FB sprint. Don't want to put any names to this but those who participated in FB sprint will know what I'm talking about. Thanks.


I might be able to do that, depending on exactly what you want. How do you measure "server load"? Do you mean the kind of information that is on the server status page? I can definitely measure a WU completion rate, but I would guess that is proportional to the GFLOPS on the server status page. Just let me know what you would like to see.

We handled the BOINC Pentathlon last year with no problems, so unless the FB sprint draws a lot more participants, I'm confident we can handle the load.
11) Message boards : Number crunching : NF is this week's Formula BOINC's Sprint (Message 2165)
Posted 7 days ago by Profile Eric Driver
Post:
we are ready for it...
12) Message boards : News : Septic Results Paper Accepted for Publication (Message 2163)
Posted 8 days ago by Profile Eric Driver
Post:
I mentioned a while back that we submitted the results of the septic search for publication. I am now happy to report that the paper has been accepted for publication in the Journal of Number Theory.

I will continue to post additional details as they become available.

Thank you to all our volunteers for making this possible!
13) Message boards : News : 2018 year in review (Message 2162)
Posted 20 days ago by Profile Eric Driver
Post:
It looks like we need several decades to finish Search 6 )


Yeah, that one is going to be difficult to finish. We will only be able to guarantee completeness up to a certain discriminant bound.
14) Message boards : Number crunching : sf6 WUs 7x8 and 9x7 missing from batch status (Message 2160)
Posted 23 days ago by Profile Eric Driver
Post:
I created a 2 dimensional representation of the batches. It can be reached via a link at the top of the batch status page or here.

This will allow the user to see every batch, including the very small ones, without cluttering up the standard tables. This representation doesn't allow for all the details like #WUs left and ETA, but then again that can be found in the standard tables (at least for the larger batches). So I think this is a good compromise.

Let me know if there are any questions.
15) Message boards : Number crunching : sf6 WUs 7x8 and 9x7 missing from batch status (Message 2159)
Posted 24 days ago by Profile Eric Driver
Post:
Good questions.

Regarding the missing entries... that was done on purpose, for no other reason than I thought the table was getting a little too big. We have been burning over 80k WUs per day, so I left out batches that I deemed were too small to be worthy. Even a case like 7x8 at 20k, was completely sent out within 5 hours and over 90% have already been returned. If you guys really want to see them I can add them to the table, its not that difficult.

For your second question... I have been doing the lower tiers of subfield 6 so that I could get some timing information for use in generating the upper tiers (i.e. to tell me how many WUs are needed). In the next couple weeks the plan is to return to subfield 5 (and DS14x11). After that, the goal is to completely close out subfield 5, which could take the better part of the year. Hopefully by then we have the GPU app working which will allow us to finally finish subfields 3 and 4.
16) Message boards : Number crunching : Cannot reach account page (Message 2156)
Posted 27 days ago by Profile Eric Driver
Post:
Sorry to hear that. This is not happening to me. I even logged out and then tried it, and it still worked.

Maybe there is something going on with cookies. Try logging out and back in. Also clear your cache and then try.

We upgraded the server OS to the latest version yesterday (about 12 hours ago), this included an upgrade of php from 7.0 to 7.2. I wonder if this could be part of the problem.
17) Message boards : News : User consent required for stats export (Message 2154)
Posted 8 Feb 2019 by Profile Eric Driver
Post:
Once consent is given how long before my stats start reporting out? Hours, days?


So I generate them once a day. I just checked and it appears to be happening every day at about 10am UTC.

I believe BOINCstats polls it once a day but it could be more often. Not sure about the other stats aggregators.

Please let us know if they are not reporting out.
18) Message boards : Number crunching : BSOD's in my last batches (Message 2152)
Posted 3 Feb 2019 by Profile Eric Driver
Post:
Sorry to hear about the BSOD. You have RAC, so are clearly crunching something, but when I query your results in the database I get nothing. Also, when I click on your user name I see the projects you are participating in, but NumberFields is not on the list. So something is fishy. I'm wondering if it's somehow related to the stats export. Did you happen to consent to stats export? This wont explain the BSOD, but some results would help to track down the problematic WUs.
19) Message boards : News : Server upgrade (Message 2150)
Posted 26 Jan 2019 by Profile Eric Driver
Post:
Looks like both problems are fixed (at least for me ...):

- BoincStats shows me in the correct country again
- In the "Projects in which you are participating list" the data for NumberFields@home is also correct

Thank you for looking into this,

Tom


Good to hear. Let me know if any other issues arise with the stats export.
20) Message boards : News : Server upgrade (Message 2147)
Posted 25 Jan 2019 by Profile Eric Driver
Post:
I was just comparing the stats files between the old db_dumps and the newer ones, and the only real difference I could see was the country tag is missing in the newer dumps. I wonder if this is the root cause of the problem. My earlier fix now puts the country into the dump file, so this may also fix the problem over at netsoft-online.


Next 20


Main page · Your account · Message boards


Copyright © 2019 Arizona State University