they added some new files to split
blc51_2bit_guppi_58543_61953_3C147_0008 thru 0025

they added some new files to split
blc51_2bit_guppi_58543_61953_3C147_0008 thru 0025

. . I had just noticed that very thing. Another new set of receiver/recorders. For some reason it is the same time period so they must have gone in at the same time. Strangely though the 4 oldest tapes are still not splitting, and 18dc09aa is still stuck :)

Stephen

< shrug >

I notice some AP being split which will help things- WU_awaiting_deletion are backing up & splitter output has fallen away for a couple of hours and the Ready_to_send buffer has dropped slightly as a result.

---

Grant
Darwin NT

I notice some AP being split which will help things- WU_awaiting_deletion are backing up & splitter output has fallen away for a couple of hours and the Ready_to_send buffer has dropped slightly as a result.

WU_awaiting_deletion continues to climb, splitter output continues to decline, as does the Ready_to_send buffer.
Will be interesting to see of the end of AP splitting allows things to clear up, or not.

---

Grant
Darwin NT

Thought I'd start a new thread for us to talk about the data files being split without messing up the panic thread. Here we can chat about what new files have been added to be split, and whatever is interesting about those files.

They just added to the list of files to be split some blc61 and blc64 for day 58543 (March 1, 2019)

Another species to add to the menagerie. Wonder what they end up crunching like?

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Another species to add to the menagerie. Wonder what they end up crunching like?

So far I have had one of the blc51(?) cpu apps take 2.5 hours. But the rest have been running under an hour. The few I have see for gpus don't seem to be crunching differently than the older formats.

Tom

---

A proud member of the OFA (Old Farts Association).

They just added to the list of files to be split some blc61 and blc64 for day 58543 (March 1, 2019)

. . Yes, I just saw them. They must have added the full complement of channel recorders, so I am expecting to see some 7x tapes soon :)

Stephen

:)

Another Arecibo file from last night being split for both MB and AP's.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Some of those BLC61s take some processing effort.

---

Grant
Darwin NT

Some of those BLC61s take some processing effort.

. . Since the majority of the bls4x-blc6x tasks have fairly consistent (and very reasonable) run times I have a growing suspicion that there is something about the 5% or so that take ridiculously long times to process that is somehow abnormal (in that the 'good' ones have ARs around 0.055 while the long one are often greater at around 0.095) and sends the crunching app into a spin. But that is for the gurus who write the apps to work out.

Stephen

? ?

Yes, some of the BLC61 cpu tasks process like a standard Arecibo AR. 90 minutes or so on the cpu.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

I just checked the Aricebo calendar, and it looks like we won't get any data recorded for Seti until late July 7/early July 8.

results received in the last hour has been creeping upward. It is now 125k. I wonder if it is just normal short WUs or noise bombs.

I'm guessing they will give us some new data on Tuesday after the outage, but the big question is, will they remove 18dc09aa??

Tune in tomorrow and find out!!!

I can't fathom any reason why they have kept 18dc09aa still plugged into the splitters. Unless somehow they have discovered that letting all the pfb splitters go idle somehow upsets the project in other manners. Unless either Eric or Jeff steps up with a tech post about why they have kept that file around, we can only guess and gripe.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

I imagine they're more focused on managing the BLC pipeline. The new technique of interlacing tasks from 41, 42, 51, 61, 64 might be aimed at reducing the risk of overwhelming the whole server farm with a continuous bombardment of noise bombs - this will give us periods of respite and recovery in between.

In that context, 18dc09aa doesn't matter. It can just sit there minding its own business until we need the Arecibo splitters again next week.

I guess if they only have one Arecibo file at a time to load, there isn't any issue with the zombie 18dc09aa file loaded as it still leaves half the AR splitters to work on any one single AR file. The question is whether if and when there is more than one Arecibo file to load that finally kicks the 18dc09aa file to the curb.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

And with half the AP splitters available it does mean that the splitting of APs takes a little longer so giving more folks a chance to enjoy them.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?