<Or maybe your original question isn't related to bottlenecks at all. Maybe you were asking how many more factors of the current active members would be needed to complete all projects immediately (without taking into account infrastructure bottlenecks and assuming unlimited capacity)? As in, "Let's throw in 10 million more volunteers at the same time and assume their IT infrastructure doesn't break....does that crunch through all the work in a day?">

Yeah, that sounds what I was thinking about. :-) The only thing I would change in your phrasing is "in a day" -> "in the longest WU runtime" (could maybe be more or less than a day). A perfectly optimized system would send the longest/heaviest WU to the most capable machines. ;-)

I have an estimation:

Assume (September 2021) around 20M wu available.
100k total users that compute 1M wu/day currently. Around 3h/result. -> 125000 wu if simultaneously
With these assumptions our theoretical structural maximum is roughly 100k times 20M/125k =
16M users. Approximate penetrating rate for this limit: .625% (a 100% penetration rate in these terms would equate to 160 times the current number of users).

Any thoughts?

UPDATE: Actually 15M and not 16M if we assume the more accurate number 3.2 hours instead of 3h. That would mean the percentage is closer to 0.67%.

It seems to me that we already overload WCG every time they release a new batch of ARP.

Mike

I understand. My calculations had the implicit assumption of the download of only one WU at a time. If we, say, download 24 work units at a time, the theoretical number of members for this hypothetical situation is much lower than 15,000,000 and is closer to 625,000 members. For current 100,000 members, the percentage would be now much higher: 16%!

Note that if each member got, on average, 6.25 other people to join our grid (with similar capacity), the theoretical number 625,000 (100%) would be reached! :-) This assumes unlimited capacity from the servers to upload all tasks at once, of course, which is obviously not the case. This and my estimations are the reason why your point is very clear to me now, dear Mike Gibson. But not sure what to think about it... :-/ I guess we surely need way, way more researchers to use our capacity (and beyond! Because 100,000 in 8B sounds like crazy small and inequity is very ingrained in this ratio of just 0.0000125). Though, to be fair, there are people who are contributing and are not in these numbers as sometimes one account is shared in the same family. Perhaps this ratio doubles to 0.000025 or more (but still, a low number).

I don't know if 20,000,000 is the current number of available WU, though... Could be higher, could be lower.