News from sub-project Deep Sieving - Page 5

firejuggler · 2016-01-23, 09:53

I had forgot 3 "0"... no wonder. i'll let it run a few hours and come back.

ET_ · 2016-01-23, 15:30

Quote:

Originally Posted by firejuggler

I had forgot 3 "0"... no wonder. i'll let it run a few hours and come back.

Wait... I incuded some optimizations to his script, and the option to create a file to insert new factors (without it, we could not parallelize the search.
You may also need the new k file already filtered up to 5T...

Hold on for a minute, I will create a new post with the environment and the instructions to work

Luigi

ET_ · 2016-11-23, 11:37

MM47: all k from 3 to 2,000,000 sieved up to 400T.

Releasing range

ET_ · 2016-11-23, 11:40

Starting sieving of MM46, k=3 to k=2,000,000.

Start 50T, target 350T. Actually sieving up to 75T.

The search can be parallelized on multiple CPUs.
Instructions on demand

Status:
http://www.doublemersennes.org/sieving/validi.php

ET_ · 2017-04-24, 14:21

Quote:

Originally Posted by ET_

Starting sieving of MM46, k=3 to k=2,000,000.

Start 50T, target 350T. Actually sieving up to 75T.

The search can be parallelized on multiple CPUs.
Instructions on demand

Status:
http://www.doublemersennes.org/sieving/validi.php

Target moved to 375T.
Actually sieved up to 261T from the bottom and 350T from the high limit (350T-375T done).
I am working with only one core (plus a spot instance of a cloud server running at 40% of its theoretical power).

ET_ · 2017-05-29, 14:59

MM36 sieved to 130T with the new Odroid-C2 PicoCluster

New targets:

MM34 sieved to 75T (Raspberry PI 3, 4 cores)
MM35 sieved to 100T (PicoCluster, 20 cores)

ET_ · 2017-06-06, 11:08

Quote:

Originally Posted by ET_

Target moved to 375T.
Actually sieved up to 261T from the bottom and 350T from the high limit (350T-375T done).
I am working with only one core (plus a spot instance of a cloud server running at 40% of its theoretical power).

MM46 : all remaining k from 3 to 2,000,000 sieved up to 375T

Range released

ET_ · 2017-06-06, 11:11

Starting sieving of MM45, k=3 to k=2,000,000.

Start 32T, target 350T.

The search can be parallelized on multiple CPUs.
Instructions on demand

Status:
http://www.doublemersennes.org/sieving/validi.php

ET_ · 2017-06-16, 07:33

Quote:

Originally Posted by ET_

MM36 sieved to 130T with the new Odroid-C2 PicoCluster

New targets:

MM34 sieved to 75T (Raspberry PI 3, 4 cores)
MM35 sieved to 100T (PicoCluster, 20 cores)

MM35 sieve to 100T completed.

ET_ · 2018-11-01, 14:52

Sieving news

MM50 reached 1000T.
MM44 reached 200T.
Other smaller ranges improved.

Many thanks to Mark Rodenkirch dmdsieve.

Meawhile our small Raspberry PI is slowly crunching MM42 through 81T->110T

ET_ · 2019-02-10, 10:57

Thank you Dylan14 for your effort, and good luck with your new project, attacking MM23, exponent 11213!

2016-11-23, 11:37	#47
ET_ Banned "Luigi" Aug 2002 Team Italia 3·1,619 Posts	MM47 - Sieve completed MM47: all k from 3 to 2,000,000 sieved up to 400T. Releasing range

2016-11-23, 11:40	#48
ET_ Banned "Luigi" Aug 2002 Team Italia 4857₁₀ Posts	MM46 - Sieve started Starting sieving of MM46, k=3 to k=2,000,000. Start 50T, target 350T. Actually sieving up to 75T. The search can be parallelized on multiple CPUs. Instructions on demand Status: http://www.doublemersennes.org/sieving/validi.php Last fiddled with by ET_ on 2016-11-23 at 11:41

2017-06-06, 11:11	#52
ET_ Banned "Luigi" Aug 2002 Team Italia 3·1,619 Posts	MM45 sieve started Starting sieving of MM45, k=3 to k=2,000,000. Start 32T, target 350T. The search can be parallelized on multiple CPUs. Instructions on demand Status: http://www.doublemersennes.org/sieving/validi.php

2018-11-01, 14:52	#54
ET_ Banned "Luigi" Aug 2002 Team Italia 3×1,619 Posts	Sieving news MM50 reached 1000T. MM44 reached 200T. Other smaller ranges improved. Many thanks to Mark Rodenkirch dmdsieve. Meawhile our small Raspberry PI is slowly crunching MM42 through 81T->110T Last fiddled with by ET_ on 2018-11-01 at 14:53

2019-02-10, 10:57	#55
ET_ Banned "Luigi" Aug 2002 Team Italia 3×1,619 Posts	MM29 - k tested to 8,000,000 Thank you Dylan14 for your effort, and good luck with your new project, attacking MM23, exponent 11213!

2016-01-23, 09:53	#45
firejuggler "Vincent" Apr 2010 Over the rainbow 2²×7×103 Posts	I had forgot 3 "0"... no wonder. i'll let it run a few hours and come back.

2017-05-29, 14:59	#50
ET_ Banned "Luigi" Aug 2002 Team Italia 12F9₁₆ Posts	MM36 sieved to 130T with the new Odroid-C2 PicoCluster New targets: MM34 sieved to 75T (Raspberry PI 3, 4 cores) MM35 sieved to 100T (PicoCluster, 20 cores)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Deep Sieving MM49 in parallel	ET_	Operazione Doppi Mersennes	22	2016-07-28 11:23
Deep Hash	diep	Math	5	2012-10-05 17:44
The news giveth, the news taketh away...	NBtarheel_33	Hardware	17	2009-05-04 15:52
Question on going deep and using cores	MercPrime	Software	22	2009-01-13 20:10
Deep Sieving 10m Digit Candidates	lavalamp	Open Projects	53	2008-12-01 03:59