Sessn49 - Nimbic's D.Gope - debugging single thread code hard, debugn parallel code *really* hard, but tools exist to help ... #48dac
2011-06-10 08:57:01Sessn49 - ExtremeDA G.Maro w/ advice - EDA tools s/b designed 4 I/O, memory [using multithread malloc, etc.], always profile - #48dac
2011-06-10 08:48:58Sessn49 - ExtremDA G.Maor - core EDA algrthms R actually simple, not that much benefit from parallelizing, using GPU or supercomputer #48dac
2011-06-10 08:46:20Sessn49 - ExtremDA G.Maor - dist system solns 5-10 yrs old, so good news - easy probs were solv'd even B 4 MCP came on stage ... #48dac
2011-06-10 08:43:51Sessn49 - okay, parallel's important & everybody agrees it'll take lotsa effort - but magic 2 make it happen not yet obvious - #48dac
2011-06-10 08:37:12Sessn49 - per TA&M's P.Li - EDA must embrace parallelism - R&D needs 2 look @ parallelizing existing code v. new parallel algrthms -- #48dac
2011-06-10 08:34:25Sessn49 - per TxA&M's P.Li notes Intel_AMD_IBM_ST/IBM_Nvidia_Dell_HP w/ MCPs - asks Y MCPs being designed w/ serial des tools? ... #48dac
2011-06-10 08:30:21Sessn49 - Per UCB's B-Y.Su, 3 main computate patterns in EDA all w/ solns - linear algebra, graph algrthms, backtrack branch&bound - #48dac
2011-06-10 08:25:20Sessn49 - Per UCB's B-Y.Su, parallel programming comes w/overhead, FLOPS R free, but mem opps R $$$ ... #48dac
2011-06-10 08:18:18Sessn49 - Per UCB's A.Vladimirescu - EDA co's should make perf data in benchmarks public to prove parallelization's a reality ... #48dac
2011-06-10 08:15:48Sessn49 - Per UCB's A.Vladimirescu, two options now - multicore wrkstatns or The Cloud - either way, thumbs up to Parallel SPICE ... #48dac
2011-06-10 08:12:13#48DAC keynote: Intel’s Gadi Singer talks about “imminent EDA transformation.” Blog report at http://bit.ly/lU77M0 #EDA #EDA360
2011-06-10 08:11:26Vladmirescu speaking - Madden already shaking his head. I suspect controversy is coming - Parallel panel #48DAC
2011-06-10 08:10:04Sessn49 - hurry to 33abd - Nanni DeMicheli's giving away his own Ferrari to whomever asks best question here in last session at DAC - #48dac
2011-06-10 08:06:30Working on post #48DAC blog - the show is not what it once was but I thought that this was a very good year - #Cadence #Denali party rocked!
2011-06-10 08:02:29Final day at #48DAC - apes, monkeys, elephants, cats & flamingos - spent the morning at the San Diego Zoo. Ready for Parallel Panel now...
2011-06-10 08:00:23RT @cadence: Here’s to another great year at #48DAC ... #CDNS will see you here again next year. Safe travels!
2011-06-10 07:41:42Sessn44 - Bio emphasis extracts single cell behavior from population data, but IC emph extracts popultn behavior from single trans .. #48dac
2011-06-10 07:26:28CF: #48DAC & the solution to the #SoC engineering glut: http://bit.ly/j1evI7
2011-06-10 07:00:55Here’s to another great year at #48DAC ... #CDNS will see you here again next year. Safe travels!
2011-06-10 07:00:38Sessn44 - Per J.Bardhan & J.Toettcher, work 2 repurpose IC CAD tools 4 Bio CAD easier said than done, but there's reason 2 hope ... #48dac
2011-06-10 06:47:47Just finished sorting & photo editing #48DAC #CDNS Denali party pics. But one favorite attached ... http://twitpic.com/599qmd
2011-06-10 06:32:23Thank you to all who visited Sonnet at #IMS2011 and #48DAC this week. Wishing you all safe travels home!
2011-06-10 06:29:54Thanks Adam! RT @SeeAdamRun: #UVM 1.1 BCL now posted: http://bit.ly/bx6D3n #48DAC fog saw user guide as BCL post. Go forth and verify! #eda
2011-06-10 06:11:39Another great year at #48DAC. We look forward to seeing you next year! #EDA360
2011-06-10 06:05:43