Getting work is sometimes hit or miss. My Windows machine is now back to a full queue after getting nothing for while. Don't know why, but glad it is back.

Cheers

Having written the previous message, after doing some groceries, one can observe the same behaviour (see my former post): there are some gaps where the distribution of workunits just pauses for a while, 15, 20 or 25 minutes.

workunit 442428181: task MCM1_0209686_7098_0  Linux         2023-12-23T14:59:18
workunit 442427181: task MCM1_0209686_6898_0  MSWin 10      2023-12-23T14:57:50
workunit 442426181: task MCM1_0209686_6701_0  Linux         2023-12-23T14:42:03
workunit 442425181: task MCM1_0209710_6256_0  MSWin 10      2023-12-23T14:40:23
workunit 442424181: task MCM1_0209688_6417_0  Linuxmint     2023-12-23T14:38:52
workunit 442423181: task MCM1_0209690_6437_0  MSWin 10      2023-12-23T14:37:45
workunit 442422181: task MCM1_0209686_5894_0  Linux openSU  2023-12-23T14:35:56
workunit 442421181: task MCM1_0209686_5697_0  MSWin 11      2023-12-23T14:34:35
workunit 442420181: task MCM1_0209690_5836_0  MSWin 10      2023-12-23T14:33:10
workunit 442419181: task MCM1_0209710_5063_0  MSWin 11      2023-12-23T14:31:43
workunit 442418181: task MCM1_0209693_5812_0  MSWin 10      2023-12-23T14:30:10

workunit 442418602: task MCM1_0209693_5897_0  Linux Ubuntu  2023-12-23T14:30:58
workunit 442417602: task MCM1_0209688_5087_0  MSWin 10      2023-12-23T14:29:25
workunit 442416602: task MCM1_0209693_5503_0  Darwin        2023-12-23T14:12:08
workunit 442415602: task MCM1_0209693_5311_0  MSWin 10      2023-12-23T14:10:40
workunit 442414602: task MCM1_0209693_5109_0  MSWin 10      2023-12-23T14:09:30
workunit 442413602: task MCM1_0209710_3953_0  MSWin 11      2023-12-23T14:07:42
workunit 442412602: task MCM1_0209686_3968_0  MSWin 10      2023-12-23T14:06:14
workunit 442411602: task MCM1_0209690_4118_0  MSWin 11      2023-12-23T14:04:48
workunit 442410602: task MCM1_0209690_3915_0  MSWin 10      2023-12-23T14:03:20
workunit 442409602: task MCM1_0209686_3367_0  MSWin 10      2023-12-23T13:41:57
workunit 442408602: task MCM1_0209690_3521_0  Fedora Linux  2023-12-23T13:40:21

Adri

Just took a look at the totals returned for MCM throughout the year. Since November 22, every succeeding day there has been just a tad fewer results returned than the day before. Over the course of that time that has been a decrease of about 162,285 work units (Nov. 22 minus Dec. 22). I believe Mike is correct in the assertion that demand for work units is exceeding supply. Probably as soon as the feeder has work units available, they are snapped up almost immediately by whichever machines are querying for work units at that time. Not knowing how the query queue is constructed this is my theory. I also don't know how fast the feeder is refilled once a batch has been exhausted, nor how many feeders may be deployed. Another factor may be the techs have slowed the release of work units to allow the backend processes to function properly - optimize operational throughput (although they seemed to be able to handle amounts in excess of a million/day in the past.)
All mere speculation on my part.

Edit: See clarification a couple of posts down. The last normal day over a million is Nov. 27 (1,055,437) and Dec 22 is (897,270) for difference of 158,167.
Sorry about the erroneous info.

Edit 2: I tried to post this correction the other day and the system crashed before I got it posted.

Cheers

Mea culpa. It is my mistake in the interpretation of my numbers. Good catch. I prepared a spreadsheet and calculated the average for each day over the previous 30 days. In my "rush to judgement" I used the wrong column - the 30 day average column - rather than the raw number column. I did this in order to smooth out the variations in the raw numbers. I will note that the raw numbers from October 22 to November 27 all show over 1 million work units returned. We have only hit that number once since that date (Dec. 5). I will continue to believe that the demand may be exceeding supply.
Thank you for bringing this to my attention.

Cheers

It does get stale, indeed.

It's the other way around for me. Although my Linux machine still has some WUs left and isn't completely dry yet.

You can add another project and set the resource share to 0, then it will only act as a filler.

I did this with Einstein@home on Nov, 30th and according to BAM my monthly stats are:

• WCG: 1,323,150 credits
  
• Einstein: 9,429,563 credits
  

Sokrates: To be is to do!
Jean-Paul Sartre: To do is to be!
Krembil: Do be do be do...