(archive)MISFIT - Page 43

Chuck · 2013-02-08, 14:17

Quote:

Originally Posted by swl551

It will give you an idea of how it will work once I account for all directories

thx

Excellent. Thanks

swl551 · 2013-02-09, 00:07

The Log purger has to write out a log indicating it purged logs.

I love it!

swl551 · 2013-02-09, 14:59

The log purger is done...

I have decided not to build the DC/LL mixer since there is low interest in DCTF and existing functionality already allows you to fetch DCTF work. It was a good idea but is expensive to build with little R.O.I.

So, do I release log purging now or wait until I have a release with a larger change set?

Chuck · 2013-02-10, 01:50

Quote:

Originally Posted by swl551

The log purger is done...

So, do I release log purging now or wait until I have a release with a larger change set?

Your call...the batch file Aramis described earlier is working for me and I can use it until the "purging" MISFIT is released (then I can delete the scheduled tasks).

What other changes are in the pipeline?

swl551 · 2013-02-10, 02:01

Quote:

Originally Posted by Chuck

Your call...the batch file Aramis described earlier is working for me and I can use it until the "purging" MISFIT is released (then I can delete the scheduled tasks).

What other changes are in the pipeline?

Currently no changes. We are reaching the functional plateau of this utility. Occasional minor tweaks, but I just don't see anything taking 100s of hours like previous builds have.

kladner · 2013-02-10, 12:56

reconsidered and removed post

swl551 · 2013-02-10, 16:59

A long standing gap in MISFIT is duplicate detection across active workers and the staged work. The lack of this feature really came to light yesterday when everyone was turning in work, hand editing all files and essentially producing the conditions where duplicates could be introduced through mass change.

In the attached screenshot you can see the new report file. It is a .CSV, already sorted by exponent and the example we have duplication both within the worker and spanning into the staged file. The red circled item is not a duplicate because its bit level is different. The report can be launched to evaluate the bit level as part of the duplicate determination or just look at exponents.

This report is not automated and is purely a diagnostic tool.
Please comment if you have any additional ideas on this specific feature.

Scott

kladner · 2013-02-10, 18:35

That looks very useful. Just open it in a spreadsheet and have a look.

swl551 · 2013-02-10, 18:42

Quote:

Originally Posted by kladner

That looks very useful. Just open it in a spreadsheet and have a look.

It puts the <---- DUPLICATE marker on the line as part of the detection. (not sure if you picked up on that).

Also the new configuration to include bit level evaluation as part of duplicate exponent detection is used in the following:
1. in the new duplicate report
2. detecting duplicates during GridStatsUpdate
3. detecting and tossing duplicates during WorkBalance

kladner · 2013-02-10, 18:47

Quote:

Originally Posted by swl551

It puts the <---- DUPLICATE marker on the line as part of the detection. (not sure if you picked up on that).

Also the new configuration to include bit level evaluation as part of duplicate exponent detection is used in the following:
1. in the new duplicate report
2. detecting duplicates during GridStatsUpdate
3. detecting and tossing duplicates during WorkBalance

Even better!

I had not realized that was part of the process.

Chuck · 2013-02-10, 19:30

I remember that there is a problem with Primenet's parsing of results if the first result submitted is a "factor found"; Primenet considers it to be a P-1 result if there are no preceeding "no factor" lines in the submission.

Does MISFIT do any reordering of results submitted to avoid this?

2013-02-09, 00:07	#464
swl551 Aug 2012 New Hampshire 2³×101 Posts	Oh the IRONY... The Log purger has to write out a log indicating it purged logs. I love it!

2013-02-09, 14:59	#465
swl551 Aug 2012 New Hampshire 2³×101 Posts	The log purger is done... The log purger is done... I have decided not to build the DC/LL mixer since there is low interest in DCTF and existing functionality already allows you to fetch DCTF work. It was a good idea but is expensive to build with little R.O.I. So, do I release log purging now or wait until I have a release with a larger change set?

2013-02-10, 12:56	#468
kladner "Kieren" Jul 2011 In My Own Galaxy! 10011110101110₂ Posts	reconsidered and removed post Last fiddled with by kladner on 2013-02-10 at 13:37 Reason: went a different route

2013-02-10, 16:59	#469
swl551 Aug 2012 New Hampshire 2³·101 Posts	Rewritten duplicate assignment detection report. A long standing gap in MISFIT is duplicate detection across active workers and the staged work. The lack of this feature really came to light yesterday when everyone was turning in work, hand editing all files and essentially producing the conditions where duplicates could be introduced through mass change. In the attached screenshot you can see the new report file. It is a .CSV, already sorted by exponent and the example we have duplication both within the worker and spanning into the staged file. The red circled item is not a duplicate because its bit level is different. The report can be launched to evaluate the bit level as part of the duplicate determination or just look at exponents. This report is not automated and is purely a diagnostic tool. Please comment if you have any additional ideas on this specific feature. Scott Attached Thumbnails Last fiddled with by swl551 on 2013-02-10 at 17:00

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
questions about misfit	jschwar313	MISFIT	1	2016-02-02 00:32
Help with Misfit	TheMawn	MISFIT	18	2015-06-21 22:00
Misfit feature request.	Aramis Wyler	MISFIT	15	2015-03-04 21:13
MISFIT 2.9	swl551	MISFIT	44	2014-12-15 12:46
MISFIT for MFAKTx	swl551	MISFIT	186	2014-02-05 23:27

2013-02-10, 18:35	#470
kladner "Kieren" Jul 2011 In My Own Galaxy! 2×3×1,693 Posts	That looks very useful. Just open it in a spreadsheet and have a look.

2013-02-10, 19:30	#473
Chuck May 2011 Orange Park, FL 3×5×59 Posts	I remember that there is a problem with Primenet's parsing of results if the first result submitted is a "factor found"; Primenet considers it to be a P-1 result if there are no preceeding "no factor" lines in the submission. Does MISFIT do any reordering of results submitted to avoid this?