Post your S5R1 completion estimates here

RandyC
RandyC
Joined: 18 Jan 05
Posts: 6,473
Credit: 111,139,797
RAC: 0
Topic 191813

P D Buck started the S4 estimation thread at more than 200 days to completion. And even though the Status Page gives an estimate of over 260 days (as of 1:40 PM UTC on Sunday, 10 September 2006), I'm going to start this thread with an estimate of
141 days.

This gives an estimated completion date of approximately 29 January 2007. To be fair, the completion date is defined as when the "Work still remaining" units counter will go to zero, not when the last WU is actually returned.

Posters are invited to give their own estimates (or countdown days) they believe the run will complete along with (optional) their rationale/explanation of their estimate.

My guestimate is based on a calculated WU completion rate of 88258 WUs/day over the last 4 days. I believe this value will fluctuate for various reasons due to students returning to classes, cooler weather in the Northern Hemisphere, (not to mention that SETI is down right now and machines have temporarily(?) switched to E@h), etc.

Any takers?

[edit: spelling]

Seti Classic Final Total: 11446 WU.

RandyC
RandyC
Joined: 18 Jan 05
Posts: 6,473
Credit: 111,139,797
RAC: 0

Post your S5R1 completion estimates here

Quote:

My guestimate is based on a calculated WU completion rate of 88258 WUs/day over the last 4 days. I believe this value will fluctuate for various reasons due to students returning to classes, cooler weather in the Northern Hemisphere, (not to mention that SETI is down right now and machines have temporarily(?) switched to E@h), etc.

Oh well, back to reality. My calculated WU completion rate has dropped back to ~71000/day. This gives an estimated time to completion of around 168 days as of 12:29 AM UTC on Sunday, 17 September 2006, and makes completion sometime around 3 March 2007. I must admit to being amazed at the effect from the temporary 'visit' of the Seti crowd last week.

I believe credit for the following graph belongs to Stef from this post. If credit belongs elsewhere, somebody please correct me.
My previous calculations were done using data collected during the Spike you see in the graph.

Seti Classic Final Total: 11446 WU.

Alinator
Alinator
Joined: 8 May 05
Posts: 927
Credit: 9,352,143
RAC: 0

The other interesting

The other interesting observation is the lack of a significant undershoot when SAH went back online. This indicates the majority of the spike was due to crunchers who use EAH as a backup project only, and implies the core body of crunchers here run EAH primarily.

Alinator

Piankhy
Piankhy
Joined: 3 May 06
Posts: 4
Credit: 417,286
RAC: 0

Here are my 2 cents: The

Message 45475 in response to message 45473

Here are my 2 cents:
The average work done in the last week (as of Sept 20th) is about 69,500 WU/day. At this rate, the remaining 11,675,575 WUs will be completed in about 168 days. Now, what could change in 168 days that would affect the throughput of this project? Would EAH succeed in attracting more users? Is there any highly-optimized EAH client in the works? How would the end of the fall semester combined with christmas holidays affect the number of active hosts? Are there more people upgrading to dual-core or even multi-core cpu's? How would the clusters (Nemo, LDAS, ...) fare during that period, are they getting upgraded? ....

Cheers!

RandyC
RandyC
Joined: 18 Jan 05
Posts: 6,473
Credit: 111,139,797
RAC: 0

It's been about 3 weeks, time

It's been about 3 weeks, time to re-evalutate the estimates...
(as of 12:30 AM UTC on Monday, 9 October 2006)

Over the past 5 days we've averaged about 72600 WUs per day. However, I've also noticed there a slight spike in the percent complete per day graph posted earlier in this thread. If we continue at this rate, I estimate completion at end-of-February 2007 or beginning of March 2007.

Does anyone have an explanation for the increase? (A lot of dual core 2's coming online to crunch???)

Seti Classic Final Total: 11446 WU.

Annika
Annika
Joined: 8 Aug 06
Posts: 720
Credit: 494,410
RAC: 0

Or maybe the need to heat

Or maybe the need to heat your room, therefore letting the boxes run 24/7 ;-)

Udo
Udo
Joined: 19 May 05
Posts: 203
Credit: 8,945,570
RAC: 0

RE: It's been about 3

Message 45478 in response to message 45476

Quote:

It's been about 3 weeks, time to re-evalutate the estimates...
(as of 12:30 AM UTC on Monday, 9 October 2006)

Over the past 5 days we've averaged about 72600 WUs per day. However, I've also noticed there a slight spike in the percent complete per day graph posted earlier in this thread. If we continue at this rate, I estimate completion at end-of-February 2007 or beginning of March 2007.

Does anyone have an explanation for the increase? (A lot of dual core 2's coming online to crunch???)

In an other thread (I don't remember which one, sorry), there was the assumption that the outage in Seti (for 7 hours or so) caused many PCs to activate the alternate project which might be E@H.
After Seti beeing available again and after processing these WUs, the number of hosts is decreasing to its 'normal' (=non vacation) value.

Udo

RandyC
RandyC
Joined: 18 Jan 05
Posts: 6,473
Credit: 111,139,797
RAC: 0

WoooHooo! Less than

WoooHooo! Less than 10,000,000 WUs left to go!!!!

Seti Classic Final Total: 11446 WU.

Annika
Annika
Joined: 8 Aug 06
Posts: 720
Credit: 494,410
RAC: 0

Not bad at all... though it

Not bad at all... though it would be interesting to know how many short and long ones that includes, because that's a really huge difference... between a bit more than one and almost twelve hours on my laptop, which would mean long WUs are about 8 times as long as short ones... certainly significant.

RandyC
RandyC
Joined: 18 Jan 05
Posts: 6,473
Credit: 111,139,797
RAC: 0

RE: Not bad at all...

Message 45481 in response to message 45480

Quote:
Not bad at all... though it would be interesting to know how many short and long ones that includes, because that's a really huge difference... between a bit more than one and almost twelve hours on my laptop, which would mean long WUs are about 8 times as long as short ones... certainly significant.

I haven't seen a short WU in ages. All 4 of my systems have only crunched long WUs for weeks (months?).

My fastest system spits them out in about 6.5-7 hrs, while my slowest system takes up to 22 hrs.

Seti Classic Final Total: 11446 WU.

Lt. Cmdr. Daze
Lt. Cmdr. Daze
Joined: 19 Apr 06
Posts: 756
Credit: 82,361
RAC: 0

RE: RE: Not bad at all...

Message 45482 in response to message 45481

Quote:
Quote:
Not bad at all... though it would be interesting to know how many short and long ones that includes, because that's a really huge difference... between a bit more than one and almost twelve hours on my laptop, which would mean long WUs are about 8 times as long as short ones... certainly significant.

I haven't seen a short WU in ages. All 4 of my systems have only crunched long WUs for weeks (months?).

My fastest system spits them out in about 6.5-7 hrs, while my slowest system takes up to 22 hrs.


I, on the other hand, haven't seen long ones in ages. I started crunching short ones first, but when I started using another Linux I got short ones ever since.

Indeed, it would be nice to know what the ratio is.

Happy crunching!

Somnio ergo sum

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.