Hi, I've released new versions of M4RI and M4RIE and I'd appreciate if someone could take the time to review the relevant tickets to get these updates into Sage:
M4RI: http://trac.sagemath.org/sage_trac/ticket/14335 M4RIE: http://trac.sagemath.org/sage_trac/ticket/14336 # What's new? # ## M4RI ## It's faster! For example. here's matrix multiplication on my computer: n 5.8 vanilla 5.8 + #14335 Magma 4096 0.09s 0.07s 0.10s 8192 0.67s 0.62s More data here: https://martinralbrecht.wordpress.com/2012/12/21/m4ri-20121224/ (20121224 never made it into Sage) ## M4RIE ## ### Performance ### It extends to degree 16 which means GF(2^e) for 10 < e <= 16 is now *a lot* faster in Sage (i.e., not embarrassingly slow) Here's multiplication for 4096 x 4096 matrices over GF(2^e) extension vanilla with patch magma 2.15-10 e: 1, 0.09s 0.07s 0.10s e: 2, 0.30s 0.22s 0.68s e: 3, 0.64s 0.46s 1.89s e: 4, 0.89s 0.70s 6.35s e: 5, 1.34s 1.05s 92.43s e: 6, 1.77s 1.39s e: 7, 2.22s 1.74s e: 8, 2.72s 2.14s e: 9, 31.21s 5.74s e: 10, 41.42s 7.94s e: 11, long 9.31s e: 12, long 10.56s e: 13, long 56.39s e: 14, long 100.64s e: 15, long 141.88s e: 16, long 199.78s 464.14s I didn't run those marked long because I ran out of RAM, check this though (without patch): sage: A = random_matrix(GF(2^11,'a'),500,500) # ~1/10 of above size sage: B = random_matrix(GF(2^11,'a'),500,500) sage: %time A*B CPU times: user 48.05 s, sys: 0.21 s, total: 48.26 s Wall time: 48.43 s Note that e>8 is not optimised yet. More information here: https://martinralbrecht.wordpress.com/2012/08/23/m4rie- support-for-finite-fields-up-to-degree-16-added/ ### Compilation ### I've split up conversion.c which means I can now do this: $ ulimit -v 196608 && make in the M4RIE directory and M4RIE successfully built with 192MB of virtual memory. Cheers, Martin -- name: Martin Albrecht _pgp: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x8EF0DC99 _otr: 47F43D1A 5D68C36F 468BAEBA 640E8856 D7951CCF _www: http://martinralbrecht.wordpress.com/ _jab: martinralbre...@jabber.ccc.de -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at http://groups.google.com/group/sage-devel?hl=en. For more options, visit https://groups.google.com/groups/opt_out.
