Posts by Vlad
log in
1) Message boards : News : This project is finally over. (Message 775)
Posted 20 Jun 2020 by Vlad
Just a tad bit late to notice lol, but I switched back to my home team evidently right after the project stopped. So now none of the Stat sites like Free-DC or BOINCstats show me on the correct team.

What would it take for this to be corrected? I'll assume the servers being brought back "online" so stats can be exported one last time?

Anyways, thanks for all that hard work Vlad...very responsive and active Admin. Need more like that!

Cheers,
bluestang

Hi bluestang,

I updated the stats on my side a couple of days ago, so your team should be correct now.
2) Message boards : News : This project is finally over. (Message 767)
Posted 6 Jul 2019 by Vlad
It would actually be nice to at least keep the BOINC contribution certificates online. I just realized that the personal account appearance has been altered dramatically and the certificate is no longer a display option.

Michael.

Sorry, I set "no_computing" option without knowing what changes it will cause. I disabled it, so the personal account pages should now appear as ususal.
3) Message boards : News : This project is finally over. (Message 762)
Posted 22 Jun 2019 by Vlad
There are few thanks to Vlad who is, (I think), the best project admin.
Always anwser, and very fast, always to help,...

Thank you again, but I just looked at some well managed projects (Amicable Numbers, RakeSearch, etc.) and tried to keep up, nothing special really. Also, a small project is much easier to manage than the big one, especially when you are the only author of the apps, which allows you to fix the bugs quickly.

Vlad, is it possible to send ONE, only ONE WU by user ?
Of course without CR. Or perhaps only 0.01.
So the problems of many request will be solved !

BOINC supports sending WU to a certain user but rather for test purposes. Such WUs do not go through validation, for example.
Yes, setting the credits to near-zero would probably solve the overload problem, but I personally find it unfair not to give the volunteers anything in return, even the credits or the joy of competition.

I understand that it is sad for the volunteers when another BOINC project finishes. But again, this project was supposed to be ended a long time ago. I only wish you could join the project in the summer of 2017 when it was in its best shape. Probably, I should have been promoting the project better in those times so that more people could join earlier.

Wish you all the best!
4) Message boards : News : 457 new structures (Message 755)
Posted 18 Jun 2019 by Vlad
Perhaps a solution : release WU hen you have. Not wait saturday.

Then it's likely that the server will be overloaded permanently, not only on Saturdays. And even if not, those who run multiple clients per host will get almost all the WUs.

The "problem" of Xanson, is only linked , or due, to the race on Formula Boinc.

I tried to contact Formula Boinc admin to ask him to exclude this project from FB but I had no opportunity to write him because he never activated my account.

Then the worst who make me sad and very disappointed. Your idea to stop and close Xanson. Once gain I understand your point of view.
I am sure in your research center, there are other project.
Not leave to fast. You can set Xanson on pause, not

So Vlad, I thank you for all your work, efforts,...
Best kind and friendly regards with hope to not read closure of Xanson

Initially, it was planned that all calculations for this project will be completed in 2017. When they were completed, I started releasing the updates each two weeks also in BOINC, because why not? But the truth is that one should not use BOINC for something that can be done on a single PC.

I got an incredibly valuable experience by running this project and I'll definitely return to BOINC if I have a valuable project. Thank you very much for your support!
5) Message boards : News : This project is finally over. (Message 751)
Posted 16 Jun 2019 by Vlad
Also, some info about the database:

The database of diffraction patterns is offline since April because the data center at the Institute for Information Transmission Problems of RAS in which it is hosted (also, this center is a hosting provider for the Acoustics@Home, USPEX@Home, SAT@Home, Optima@Home, Boinc.ru and some other non-boinc resources) is powered off and in the middle of moving to a new location. As soon as the data center will be powered on, the database will be back online. However, moving the academic data center to a new location is a long process because a lot of bureaucracy involved.
6) Message boards : News : This project is finally over. (Message 750)
Posted 16 Jun 2019 by Vlad
The inability of the project to provide sufficient number of WUs for everyone provokes major server overloads during the WU generation each two weeks. These overloads significantly complicate downloading/uploading of the work. Continuing the project will cause an unpleasant experience for the participants.

Considering that the power of BOINC is actually not required for this project since the project reached its goal in September 2017, I think this is a right moment to finally end XANSONS for COD.

I would like to thank everyone for helping the project all that time!
7) Message boards : News : 457 new structures (Message 749)
Posted 16 Jun 2019 by Vlad
First of all, thanks to everyone for sharing their ideas. With some help from participants I found out the following:

1) The number of requests per seconds causing the overload is hundreds, not thousands. However, hundreds of requests/s is also a big number for a single server (large projects usually have multiple servers and the load balancer).

2) The min_sendwork_interval parameter sets the time out delay even if the work was not sent to the host, so I was wrong about that. This parameter is set to 7 seconds for this project. However, increasing it will not solve the problem (see 3).

3) One can easily overcome the limitation set by the min_sendwork_interval parameter (as well as the limitation on maximum tasks in progress) by launching multiple clients on a single host. In this case each client will be treated as a separate host with its unique ID. The parameter multiple_clients_per_host, which could prohibit the scheduler from registering each client as a separate host, is deprecated since 2011. The huge number of requests, which are causing the overload, come not from different hosts but from many different clients running on the same hosts. And this is also the reason why most of the tasks are processed by only a few users.

4) Surely, the scheduler could be modified to distinguish the unique hosts from just different versions of clients running on the same host by the combination of local IP, domain name and the external IP. However, one can set up a proxy for each client to make the external IPs unique. But even if the reliable method of finding the unique hosts exist, it would require querying over the hundreds of the hosts in the database, which will significantly slow down the scheduler.

So, the whole situation with the overload is a result of inability of this project to provide sufficient number of WUs for everyone. I do not blame in any way the users who try to get as much WUs as possible by using various methods, because I understand that the competition is a major part of BOINC. However, continuing the project will cause painful experience for the participants and this is definitely not what I want. Considering that the power of BOINC is actually not required for this project, since the project reached its goal in September 2017, I think this is a right moment to end XANSONS for COD. I’ll announce the finish of the project in a separate thread.
8) Message boards : News : 457 new structures (Message 744)
Posted 9 Jun 2019 by Vlad
I think the BOINC forums mentioned minimum was 6 seconds.

You are probably talking about the "min_sendwork_interval" parameter. If I understand correctly, this parameter tells the client not to request new WUs earlier than X seconds after receiving the last one. But if the client has no work at all, it can request it as often as the client configuration allows to.
Thanks for the suggestions anyway.
9) Message boards : News : 457 new structures (Message 743)
Posted 9 Jun 2019 by Vlad
3.000 host for 30.000 request/sec..... In the worst case : each user
need to make 10 request by second !!!.........

Sorry, I already made the correction in my post. It is actually 30.000 packeges/sec (this is what the AWS statistics says). I do not know how many packages are in a single work request but I'll try to find it out.
10) Message boards : News : 457 new structures (Message 739)
Posted 8 Jun 2019 by Vlad
Ok, temporary server upgrade didn’t help. The limitation seems to be in the number of incoming requests per second (about 30,000 packages/s) and not in the Mbit/s. Sorry, I don’t know how to overcome this for now.
11) Message boards : News : 457 new structures (Message 738)
Posted 8 Jun 2019 by Vlad
New COD entries are added.

Update. All WUs are sent to the hosts.
12) Message boards : News : 691 new structures (Message 737)
Posted 8 Jun 2019 by Vlad
Hello Vlad,
I hope you could adapt network !
If I look new registred users : 0
But new hosts in the past 24 hours : 207 !!!
It will say from the 141 recent users some have add a lot of hosts .
It will be very big bottleneck !
It increase release after release...

Ii will be a very hot saturday evening for your server !!!

I upgraded the virtual server performance to the same level as it was during the last weeks of active phase of the project in September 2017. So, the network bandwidth is about 500 Mbit/s now. However, it seems that limitations come from the number of incoming packages per second and unfortunately this parameter cannot be increased by simple server upgrade.
13) Message boards : News : 691 new structures (Message 734)
Posted 27 May 2019 by Vlad
That was partly my fault. I was trying to complete my 100 hours for WUProps on the CPU and GPU work units and the work here is so rare. I was manually updating the project and retrying stalled d/l-u/l's for 29 machines Saturday for 3 hours before work. It was hectic and like real work...

Have 250 hours on the CPU now (would have had it 2 weeks prior but 24 machines were missing proper C++ libraries) and the 100 on GPU should be done by my slow ATI HD 7450M by tomorrow.

I'll let the BOINC client work it's slow methodical way from now on.

That is another thought; does the feeder have a parameter to slow the WU generation to 1 per minute so about 13 hours for a batch like yesterday?

No, this is nobody’s fault. It is quite natural that the users want to get as much WUs as possible under the conditions where the number of WUs is highly limited. I think that slowing down the WU generation will make the process even more painful for those users who stop other projects to speed up the calculations for this one. I hope that temporarily server upgrade will help with the network bandwidth and no other actions will be required.
14) Message boards : News : 691 new structures (Message 730)
Posted 26 May 2019 by Vlad
Yep same, I get a handful of WU's and then the uploads & downloads lock up completely. The problem is only with xansons as all other boinc work and internet activities remain unaltered, then 4 or 5 hours later the WU's finally finish going where there supposed too, by then all work is gone.
According to the server that's from just 118 users. Actually probably a lot less users than that as many users download both CPU and GPU work at the same time. Problem is it seems to be getting worse each fortnight.

@ Vlad: I realize this project is just running part time now but is there any way of increasing bandwidth on the day? I mean it's only needed for at most 6 hours once a fortnight.
It such a short time period maybe your ISP would do you a solid and give you a bit of bandwidth flexibility once a fortnight?
Stranger things have happened. LOL

All the best.

Yes, sorry for that. It's almost like a DDOS for the server each time and it's getting worse. I think I can increase network bandwidth by a factor of 3 or 4 by temporarily upgrading the virtual server. I'll do it by the next time, let's see if it helps.
15) Message boards : News : 691 new structures (Message 727)
Posted 25 May 2019 by Vlad
New COD entries are added.

Update. All WUs are sent to the hosts.
16) Message boards : News : 495 new structures (Message 726)
Posted 11 May 2019 by Vlad
New COD entries are added.

Update. All WUs are sent to the hosts.
17) Message boards : News : 455 new structures (Message 724)
Posted 30 Apr 2019 by Vlad
How many tasks are generated by 455 new structures? Just curious.

I suppose a better question is, how many tasks does 1 new structure generate?

In general, 36 WUs are generated per one structure (18 for the x-ray diffraction and 18 for the neutron diffraction). Usually, only about 20% of them are sent to CPU, but this number depends on the current ratio of unsent WUs in the CPU and GPU queues.
18) Message boards : News : 455 new structures (Message 722)
Posted 29 Apr 2019 by Vlad
Are those WUs were for GPU only?

No, but the number of GPU tasks was higher than that for the CPU.
19) Message boards : News : 455 new structures (Message 720)
Posted 27 Apr 2019 by Vlad
New COD entries are added.

Update. All WUs are sent to the hosts.
20) Message boards : News : 428 new structures (Message 719)
Posted 13 Apr 2019 by Vlad
New COD entries are added.

Update. All WUs are sent to the hosts.


Next 20

Main page · Your account · Message boards


© 2020 Vladislav Neverov (NRC 'Kurchatov institute'), Nikolay Khrapov (Institute for Information Transmission Problems of RAS)