Meta tags:
Author=Rich Vuduc;
keywords=Berkeley, benchmarking, performance tuning, BeBOP;
description=Berkeley Benchmarking and Optimization Home Page;

Headings (most frequently used words):

projects, 2008, 2005, 2004, 2003, 2002, people, software, distributions, awards, publications, talks, and, posters, related, principal, investigators, affiliated, researchers, graduate, students, previous, participants, 2017, 2016, 2015, 2014, 2013, 2012, 2011, 2010, 2009, 2007, 2006, 2001, current, local, activities, prior, external, collaborations,

Text of the page (most frequently used words):
and (226), pdf (183), demmel (134), james (133), paper (114), for (76), the (70), communication (68), eecs (67), ucb (66), parallel (58), matrix (57), report (57), yelick (50), katherine (49), 2013 (49), tech (47), 2012 (43), sparse (41), 2014 (41), grey (36), ballard (36), performance (35), avoiding (33), talk (31), oded (29), schwartz (29), 2015 (29), international (28), ieee (28), see (28), also (27), 2008 (26), symposium (26), berkeley (25), proceedings (24), science (23), may (23), tuning (23), computing (23), vuduc (23), vector (22), processing (22), richard (22), edgar (22), solomonik (22), conference (21), algebra (21), multiplication (21), siam (21), 2011 (21), slides (20), computer (20), algorithms (20), oliker (20), erin (20), carson (20), linear (19), distributed (19), acm (18), leonid (18), mark (17), ipdps (17), 2017 (17), optimal (16), november (16), knight (16), multiply (15), june (15), williams (15), supercomputing (15), john (15), laura (15), grigori (15), 2009 (15), best (14), usa (14), 2004 (14), architectures (14), nicholas (14), automatic (13), kernels (13), 2003 (13), march (13), bounds (13), benjamin (13), thesis (13), michael (13), journal (13), august (13), award (13), olga (13), holtz (13), aydın (13), buluç (13), powerpoint (12), scientific (12), hoemmen (12), postscript (12), 2005 (12), shoaib (12), kamil (12), division (12), samuel (12), shalf (12), 2007 (12), david (12), 2010 (12), lipshitz (12), with (11), 2002 (11), symmetric (11), computations (11), gzip (11), methods (11), optimization (11), analysis (11), spaa (11), evangelos (11), georganas (11), nguyen (11), yang (11), 2016 (11), technical (10), high (10), algorithm (10), february (10), krylov (10), factorization (10), you (10), numerical (9), december (9), using (9), parallelism (9), prize (9), july (9), hong (9), diep (9), rajesh (8), nishtala (8), workshop (8), subspace (8), fast (8), version (8), applications (8), october (8), multicore (8), data (8), sivan (8), toledo (8), software (7), kaushik (7), datta (7), stencil (7), april (7), jin (7), lee (7), framework (7), andrew (7), step (7), penporn (7), koanantakool (7), genome (7), appear (6), energy (6), san (6), california (6), efficient (6), matrices (6), ppt (6), marghoob (6), mohiyuddin (6), abstract (6), jeff (6), lanczos (6), dongarra (6), 2006 (6), engineering (6), vasily (6), volkov (6), proc (6), lower (6), gearhart (6), accuracy (6), aditya (6), devarakonda (6), lab (6), talks (5), research (5), from (5), information (5), meeting (5), jan (5), francisco (5), optimizations (5), atlanta (5), georgia (5), computational (5), jack (5), systems (5), transactions (5), par (5), pivoting (5), tensor (5), scaling (5), reproducible (5), extreme (5), razvan (5), carbunescu (5), steven (5), ben (5), alex (5), tall (5), skinny (5), householder (5), anal (5), appl (5), method (5), lbnl (5), papers (4), this (4), publications (4), are (4), work (4), copyright (4), hpc (4), benchmark (4), scale (4), optimizing (4), sparsity (4), precision (4), poster (4), implementation (4), collective (4), multiple (4), cache (4), oski (4), automatically (4), kernel (4), hyun (4), 2001 (4), models (4), eun (4), hsu (4), based (4), winner (4), september (4), exploiting (4), benchmarking (4), platforms (4), gpus (4), cholesky (4), factorizations (4), optimized (4), auto (4), hua (4), xiang (4), dense (4), sequential (4), decomposition (4), point (4), minimizing (4), kurt (4), keutzer (4), siag (4), graph (4), costs (4), ming (4), rank (4), revealing (4), springer (4), eliahu (4), armando (4), fox (4), omer (4), spillinger (4), invited (4), trans (4), arithmetic (4), discovery (4), environment (4), xsede (4), alameda (4), gordon (4), dulceneia (4), becker (4), druinsky (4), inon (4), peled (4), ichitaro (4), yamazaki (4), mat (4), between (4), jarrod (4), chapman (4), daniel (4), rokhsar (4), arxiv (4), under (3), national (3), intel (3), presented (3), all (3), works (3), solvers (3), project (3), extended (3), projects (3), related (3), given (3), scidac (3), mpi (3), hardware (3), communications (3), state (3), art (3), blocking (3), tuned (3), libraries (3), sam (3), moon (3), bilmes (3), via (3), memory (3), block (3), modeling (3), structure (3), parry (3), husbands (3), hormozd (3), gahvari (3), january (3), programming (3), jonathan (3), carter (3), ankit (3), jain (3), murphy (3), computation (3), ali (3), group (3), published (3), sisc (3), 131 (3), euro (3), distinguished (3), expansion (3), improving (3), lecture (3), notes (3), practice (3), successive (3), band (3), reduction (3), hammond (3), gonzález (3), 126 (3), strong (3), annual (3), strassen (3), recursive (3), rectangular (3), storage (3), highlight (3), scalable (3), 27th (3), floating (3), summation (3), driscoll (3), direct (3), maryam (3), mehri (3), dehnavi (3), that (3), jay (3), susan (3), mehringer (3), big (3), mathias (3), jacquelin (3), 175 (3), reconstructing (3), vectors (3), brian (3), university (3), novo (3), eth (3), zurich (3), support (3), machines (3), czechowski (3), song (3), ariful (3), azad (3), learning (3), election (3), academy (3), memorial (3), these (2), year (2), preparation (2), supported (2), part (2), nsf (2), cooperative (2), agreement (2), aci (2), not (2), authors (2), most (2), effort (2), tops (2), self (2), adapting (2), system (2), phipac (2), blas (2), parlab (2), presentation (2), recent (2), google (2), day (2), our (2), approach (2), triangular (2), solve (2), jpeg (2), gif (2), png (2), cse (2), selection (2), finding (2), triple (2), link (2), 420k (2), register (2), blocked (2), copenhagen (2), denmark (2), 3mb (2), 138k (2), para (2), when (2), why (2), library (2), future (2), codes (2), lapack (2), bandwidth (2), posters (2), statistical (2), jen (2), level (2), attila (2), gyulassy (2), mathematics (2), size (2), lorimier (2), lijue (2), zhong (2), csd (2), 950k (2), evaluation (2), italy (2), cell (2), processor (2), master (2), austin (2), minutes (2), 560k (2), leading (2), poski (2), autotuning (2), networks (2), gaussian (2), elimination (2), pinar (2), sc09 (2), methodology (2), power (2), pptx (2), one (2), sided (2), brief (2), announcement (2), model (2), 228k (2), lenny (2), amal (2), khabou (2), panel (2), its (2), cyclops (2), jorge (2), domínguez (2), yili (2), zheng (2), juan (2), touriño (2), overlapping (2), perfect (2), additional (2), consumption (2), 24th (2), 197 (2), residual (2), replacement (2), strategy (2), maximum (2), attainable (2), 205 (2), networking (2), dec (2), appeared (2), cacm (2), jong (2), byun (2), imaging (2), runtime (2), sci (2), comput (2), julien (2), langou (2), devin (2), matthews (2), massively (2), 21st (2), column (2), ppam (2), powers (2), body (2), course (2), 127 (2), indefinite (2), 25th (2), layout (2), intern (2), conf (2), cindy (2), rubio (2), bailey (2), tradeoffs (2), synchronization (2), finite (2), gradient (2), van (2), straalen (2), nick (2), contraction (2), comp (2), architecting (2), autograder (2), code (2), 165 (2), eigenproblem (2), multimedia (2), assembly (2), solver (2), kenneth (2), clusters (2), doi (2), theory (2), hofmeyr (2), rob (2), egan (2), mahoney (2), coordinate (2), descent (2), liu (2), 30th (2), dmitriy (2), morozov (2), sang (2), yun (2), cho (2), jui (2), rich (2), design (2), imagenet (2), training (2), 1710 (2), marquita (2), ellis (2), brandon (2), sciences (2), years (2), george (2), fellowship (2), fellow (2), american (2), awards (2), interface (2), distributions (2), people (2), application (2), among (2), interaction (2), bebop (2), reports, writing, delivered, conjunction, was, foundation, 9813362, 9619020, department, doe, grant, fc02, 01er25478, gift, here, does, necessarily, reflect, position, policy, government, official, endorsement, should, inferred, standard, disclaimer, ensure, timely, dissemination, scholarly, rights, therein, retained, other, holders, persons, copying, expected, adhere, terms, constraints, invoked, each, author, cases, reposted, without, explicit, permission, holder, challenge, tera, pde, sans, external, collaborations, java, compiler, titanium, intelligent, ram, iram, toolkit, mixed, xblas, prior, upc, degas, sejits, aspire, current, local, activities, kick, off, describes, results, discussion, ideas, acceleration, pagerank, 139k, pics, 142, 830k, 806, shown, stanford, summarizes, overall, number, including, non, 715, 520, 617k, bay, area, overview, probabilistic, pass, choosing, operation, platform, full, 118k, 985k, 436k, 504k, 177k, 678k, computationally, product, class, schur, complement, 331k, 319k, seminar, adaptable, benchmarks, 147k, orlando, florida, 305k, next, scalapack, 952k, implementations, akx, 05mb, 45mb, 7mb, 940k, microsoft, visit, draft, splitting, reordering, 685k, 163k, 410k, 132k, 164k, reuse, new, york, 681k, 548k, languages, ics, baltimore, 159k, 834k, melbourne, australia, 1232, 735k, 91k, 328k, iccs, hierarchy, senior, dept, christopher, 273k, word, effects, 867k, 1297, 983k, empirical, search, 135, 158, 8mb, 1335, ismail, bustany, cleve, ashcraft, toward, products, montreal, quebec, canada, 540k, 204k, 178k, victor, eijkhout, erika, fuentes, antoine, petitet, clint, whaley, 600k, special, issue, program, generation, adaptation, 190k, physics, series, sorrento, 322k, variable, blocks, ischia, 216k, frontiers, potential, jose, 604k, correctness, implicit, explict, spec, five, 390k, applicable, 376k, 123, subspaces, 438k, emerging, 179, accelerate, bisection, eigenvalues, tridiagonal, track, lattice, boltzmann, simulation, capabilities, masters, extensible, perform, spmvs, kubiatowicz, 282k, embedding, hybrid, electric, photonic, tiled, cmps, patterson, 612k, 584k, student, 648k, tune, review, modern, microprocessors, 400kb, hotpar, multicores, preprint, 8467k, tpds, requirements, interconnect, end, chan, howison, wes, bethel, prabhat, 354k, cray, user, generalized, 152k, 464k, iwapt2009, fourth, kathy, wawrzynek, domain, 428k, calu, pierre, yves, sylvain, peyronnet, problem, anderson, ioana, dumitriu, 852k, eigenproblems, singular, value, 463k, updated, 700k, 791k, reduced, multithreaded, nov, abhinav, bhatele, topology, aware, collectives, 793k, two, 689k, vecpar, lcns, kobe, japan, multidimensional, torus, 1357k, principles, ppopp, preliminary, 168, instrumenting, chip, counters, 248k, heterogeneous, independent, 520k, 7659, 194, 1st, mediterranean, medalg, lin, 2223k, 215, medical, 1250, 1262, marc, alley, shreyas, vasanawala, lustig, spirit, compressed, sensing, mri, clinically, feasible, a206, a239, pairs, shortest, paths, reducing, eliminating, load, imbalance, contractions, reproducibility, exascale, lncs, interactions, magnetics, dennis, giannacopoulos, yousef, kurdi, techniques, graphics, units, christ, thomas, scanlon, original, revised, programs, reference, arrays, providing, online, complexity, bound, rational, functions, gerhart, implementing, aasen, dynamic, scheduler, 222, 231, random, 232, 241, partial, shape, morphing, bigdata, oct, santa, clara, benson, gleich, mapreduce, gonzalez, cuong, william, kahan, koushik, sen, costin, iancu, hough, precimonious, assistant, s42, s61, nonsymmetric, comunications, 107, 114, feb, biconjugate, error, mike, lijewski, ann, almgren, bottom, geometric, multigrid, acta, numerica, cambridge, press, 155, etna, electronic, 125, 141, deflation, technique, conjugate, stanton, 143, coupled, cluster, 26th, csc, 6th, combinatorial, 147, contention, combinations, graphs, network, topologies, provably, 10th, applied, wyrzykowski, karczewski, waniewski, eds, berlin, heidelberg, 2mb, detection, tomm, ekaterina, gonina, gerald, friedland, eric, battenberg, content, python, bruijn, construction, traversal, contracting, tensors, fewer, multiplications, torsten, hoefler, bidiagonalization, routine, associated, least, squares, frpa, svm, 22nd, symp, arith, society, jun, destefano, rivera, extending, access, skills, through, blended, line, computers, 1109, 2345391, 793, 819, distr, aug, available, org, 1016, jpdc, 003, loop, nests, willow, ahrens, 229, biology, martin, mascher, kerrie, barry, adam, session, veronika, strnadova, jerry, jenkins, sunish, sehgal, jeremy, schmutz, uwe, scholz, robbie, waugh, jesse, poland, gary, muehlbauer, nils, stein, whole, shotgun, assembling, anchoring, hexaploid, bread, wheat, 29th, meraligner, fully, sequence, aligner, chaitanya, aluru, hipmer, assembler, 1612, 04003, kimon, fountoulakis, primal, dual, gittens, evan, racah, ringenburg, lisa, gerhardt, jey, kottalam, jialin, kristyn, maschhoff, shane, canon, jatin, chhugani, pramod, sharma, jiyan, jim, harrell, venkat, krishnamurthy, comparison, analytics, spark, three, case, studies, harsha, vardhan, simhadri, 163, write, neural, xiangru, lian, hsiang, inderjit, dhillon, hsieh, asynchronous, greedy, kent, classifier, c624, c651, levels, 1709, 05011, zhao, zhang, hseih, 156, igor, gitman, boris, ginsburg, sgd, batch, 32k, 08883, saeed, soori, mert, gurbuzbalaban, proximal, convex, problems, 10769, alnur, massive, graphical, cook, characterization, prallel, scheduling, machine, dataset, deep, gpu, knights, landing, download, bibtex, entries, icpp, sidney, fernbach, ramamoorthy, charles, babbage, athena, lecturer, doctoral, dissertation, honorable, mention, sakrison, selected, inclusion, influential, 1987, out, about, 1800, total, krste, asanovic, chee, whye, chin, portable, ansi, coding, paris, kanellakis, association, advancement, induction, county, women, hall, fame, arts, finalist, spmv, utility, converting, file, formats, matrixmarket, harwell, boeing, converter, collection, low, primitives, provide, use, basic, subprograms, reproblas, wei, thompson, randy, shoopman, geoff, pike, tuyet, linh, phan, jaeseop, gina, kim, suh, kang, chris, yozo, hida, gaeke, amik, singh, previous, participants, bruns, smith, brock, grace, dinh, rebecca, roelofs, orianna, demasi, graduate, students, switzerland, sandia, hebrew, xiaoye, kookmin, adams, davis, mit, affiliated, researchers, principal, investigators, resources, meetings, reading, links, below, detailed, status, automating, process, starting, which, dominate, retrieval, managing, trade, offs, various, measures, such, speed, compilers, broadly, interested, understanding, issues, implications, general, interests, home, page,

Text of the page (random words):
alex druinsky inon peled oded schwartz sivan toledo ichitaro yamazaki pdf paper perfect strong scaling using no additional energy proceedings of the ieee international parallel distributed processing symposium ipdps 13 2013 james demmel andrew gerhart ben lipshitz and oded schwartz pdf see also ucb eecs 2012 126 paper communication optimal parallel recursive rectangular matrix multiplication proceedings of the ieee international parallel distributed processing symposium ipdps 13 2013 james demmel david eliahu armando fox shoaib kamil ben lipshitz oded schwartz and omer spillinger pdf see also ucb eecs 2012 205 ph d thesis avoiding communication in dense linear algebra computer science division u c berkeley august 2013 grey ballard pdf tech report avoiding communication in successive band reduction eecs technical report no ucb eecs 2013 131 july 11 2013 grey ballard james demmel nicholas knight pdf tech report an arithmetic complexity lower bound for computing rational functions with applications to linear algebra eecs technical report no ucb eecs 2013 126 july 1 2013 james demmel pdf paper providing a supported online course on parallel computing extended abstract proc of conference on extreme science and engineering discovery environment xsede 13 july 2013 jay alameda razvan carbunescu james demmel steven gordon and susan mehringer pdf tech report communication lower bounds and optimal algorithms for programs that reference arrays part 1 eecs technical report no ucb eecs 2013 61 may 14 2013 michael christ james demmel nicholas knight thomas scanlon and katherine yelick pdf revised version pdf original paper communication avoiding krylov techniques on graphics processing units ieee trans on magnetics v 49 n 5 may 2013 maryam mehri dehnavi james demmel dennis giannacopoulos yousef el kurdi pdf paper a communication optimal n body algorithm for direct interactions in proceedings of 27th ieee international parallel distributed processing symposium ipdps 13 2013 michael driscoll evangelos georganas penporn koanantakool edgar solomonik and katherine yelick pdf tech report exploiting data sparsity in parallel matrix powers computations eecs technical report no ucb eecs 2013 47 may 3 2013 erin carson james demmel nicholas knight pdf to appear in proceedings of ppam 2013 springer lncs tech report communication avoiding rank revealing qr factorization with column pivoting eecs technical report no ucb eecs 2013 46 james demmel laura grigori ming gu and hua xiang pdf paper fast reproducible floating point summation 21st ieee symposium on computer arithmetic april 2013 james demmel and hong diep nguyen pdf paper numerical accuracy and reproducibility at exascale 21st ieee symposium on computer arithmetic april 2013 james demmel and hong diep nguyen pdf paper cyclops tensor framework reducing communication and eliminating load imbalance in massively parallel contractions in proceedings of 27th ieee international parallel distributed processing symposium ipdps 13 2013 edgar solomonik devin matthews jeff hammond and james demmel pdf see also ucb eecs 2013 11 paper minimizing communication in all pairs shortest paths in proceedings of 27th ieee international parallel distributed processing symposium ipdps 13 2013 edgar solomonik aydın buluç and james demmel pdf see also ucb eecs 2013 10 2012 paper communication optimal parallel and sequential qr and lu factorizations siam j sci comput 34 1 a206 a239 february 2012 james demmel laura grigori mark hoemmen and julien langou siag on supercomputing best paper prize pdf see also ucb eecs 2008 89 paper fast l1 spirit compressed sensing parallel imaging mri scalable parallel implementation and clinically feasible runtime ieee trans medical imaging v 31 i 6 pp 1250 1262 2012 marc alley mark murphy kurt keutzer shreyas vasanawala james demmel and michael lustig pdf tech report autotuning sparse matrix vector multiplication for multicore ucb eecs 2012 215 november 2012 jong ho byun richard lin katherine a yelick and james demmel pdf 2223k paper graph expansion and communication costs of fast matrix multiplication in journal of the acm 2012 grey ballard james demmel olga holtz and oded schwartz invited to appear as cacm research highlight pdf 228k paper graph expansion analysis for communication costs of fast rectangular matrix multiplication proceedings of the 1st mediterranean conference on algorithms medalg 12 dec 2012 appeared in lecture notes in computer science springer v 7659 pp 13 36 grey ballard james demmel olga holtz benjamin lipshitz oded schwartz pdf see also ucb eecs 2012 194 paper communication avoiding and overlapping for numerical linear algebra november 2012 in proceedings of 2012 international conference for high performance computing networking storage and analysis sc 12 evangelos georganas jorge gonzález domínguez edgar solomonik yili zheng juan touriño and katherine yelick pdf 520k paper communication avoiding parallel strassen implementation and performance november 2012 in proceedings of 2012 international conference for high performance computing networking storage and analysis sc 12 grey ballard james demmel benjamin lipshitz and oded schwartz pdf see also ucb eecs 2012 90 tech report communication optimal parallel recursive rectangular matrix multiplication ucb eecs 2012 205 james demmel david eliahu armando fox shoaib kamil benjamin lipshitz oded schwartz omer spillinger pdf tech report a residual replacement strategy for improving the maximum attainable accuracy of s step krylov subspace methods ucb eecs 2012 197 september 2012 erin carson and james demmel pdf paper communication optimal parallel algorithm for strassen s matrix multiplication proceedings of the 24th annual acm symposium on parallelism in algorithms and architectures spaa 12 june 2012 grey ballard james demmel olga holtz benjamin lipshitz oded schwartz pdf see also ucb eecs 2012 32 paper strong scaling of matrix multiplication algorithms and memory independent communication lower bounds brief announcement proceedings of the 24th annual acm symposium on parallelism in algorithms and architectures spaa 12 june 2012 grey ballard james demmel olga holtz benjamin lipshitz and oded schwartz pdf see also ucb eecs 2012 31 tech report communication bounds for heterogeneous architectures june 2012 grey ballard james demmel and andrew gearhart pdf 248k tech report instrumenting linear algebra energy consumption via on chip energy counters ucb eecs 2012 168 june 2012 james demmel and andrew gearhart pdf tech report perfect strong scaling using no additional energy ucb eecs 2012 126 may 2012 james demmel andrew gearhart oded schwartz and benjamin lipshitz pdf tech report communication avoiding and overlapping for numerical linear algebra ucb eecs 2012 65 may 2012 evangelos georganas jorge gonzález domínguez edgar solomonik yili zheng juan touriño and katherine yelick pdf tech report sequential communication bounds for fast linear algebra ucb eecs 2012 36 march 2012 grey ballard james demmel olga holtz oded schwartz pdf tech report a preliminary analysis of cyclops tensor framework ucb eecs 2012 29 march 2012 edgar solomonik jeff hammond and james demmel pdf paper communication avoiding successive band reduction acm symposium on principles and practice of parallel programming ppopp february 2012 grey ballard nicholas knight and james demmel pdf 1357k tech report matrix multiplication on multidimensional torus networks vecpar 12 springer lecture notes in computer science lcns kobe japan july 2012 edgar solomonik and james demmel pdf see also ucb eecs 2012 28 tech report lu factorization with panel rank revealing pivoting and its communication avoiding version ucb eecs 2012 15 january 2012 amal khabou james demmel laura grigori and ming gu pdf 689k 2011 tech report avoiding communication in two sided krylov subspace methods ucb eecs 2011 93 august 2011 erin carson nicholas knight and james demmel pdf 793k paper improving communication performance in dense linear algebra via topology aware collectives supercomputing 11 nov 2011 edgar solomonik abhinav bhatele and james demmel pdf see also ucb eecs 2011 92 paper reduced bandwidth multithreaded algorithms for sparse matrix vector multiplication international parallel distributed processing symposium ipdps 11 may 2011 aydın buluç samuel williams lenny oliker james demmel pdf 791k paper graph expansion and communication costs of fast matrix multiplication acm symposium on parallelism in algorithms and architectures spaa 11 june 2011 grey ballard james demmel olga holtz and oded schwartz spaa best paper award pdf 228k tech report communication optimal parallel 2 5d matrix multiplication and lu factorization algorithms ucb eecs 2011 10 february 2011 edgar solomonik and james demmel euro par distinguished paper award pdf 700k tech report minimizing communication in numerical linear algebra ucb eecs 2011 15 february 2011 grey ballard james demmel olga holtz oded schwartz siam siag la prize pdf 560k 2011 updated version 2010 paper communication optimal parallel and sequential cholesky decomposition siam journal on scientific computing sisc december 2010 grey ballard james demmel olga holtz and oded schwartz pdf 463k conference version published in spaa 2009 tech report minimizing communication for eigenproblems and the singular value decomposition ucb eecs 2011 14 february 2011 grey ballard james demmel and ioana dumitriu pdf 852k tech report communication avoiding qr decomposition for gpus ucb eecs 2010 131 october 2010 michael anderson grey ballard james demmel and kurt keutzer pdf 2 3m paper brief announcement lower bounds on communication for sparse cholesky factorization of a model problem symposium on parallelism in algorithms and architectures spaa 2010 june 2010 laura grigori pierre yves david james demmel sylvain peyronnet pdf 99 5k tech report calu a communication optimal lu factorization algorithm ucb eecs 2010 29 march 2010 james demmel laura grigori hua xiang pdf 428k ph d thesis communication avoiding krylov subspace methods computer science division u c berkeley may 2010 mark hoemmen pdf 4 2m 2009 ph d thesis auto tuning stencil codes for cache based multicore platforms computer science division u c berkeley december 2009 kaushik datta pdf 3 1m talk slides pptx 4 1m or pdf 3 3m ph d thesis automatically tuning collective communication for one sided programming models computer science division u c berkeley december 2009 rajesh nishtala pdf 4 7m talk slides pptx 5 9m or pdf 8 2m paper a methodology for domain optimized power efficient supercomputing in proc of sc09 november 2009 marghoob mohiyuddin mark murphy leonid oliker john shalf john wawrzynek and sam williams pdf 950k talk slides pdf 3 3m paper minimizing communication in sparse matrix solvers in proc of sc09 november 2009 marghoob mohiyuddin mark hoemmen james demmel and kathy yelick pdf 1 2m talk slides pdf 3 9m paper auto tuning the 27 point stencil for multicore iwapt2009 the fourth international workshop on automatic performance tuning october 2009 kaushik datta samuel williams vasily volkov jonathan carter leonid oliker john shalf and katherine yelick pdf 464k talk slides pdf 5 7 mb paper communication optimal parallel and sequential cholesky decomposition symposium on parallelism in algorithms and architectures spaa 2009 august 2009 grey ballard james demmel olga holtz and oded schwartz pdf 152k journal version published in sisc 2010 paper a generalized framework for auto tuning stencil computations cray user group conference atlanta ga may 2009 winner best paper shoaib kamil cy chan sam williams leonid oliker john shalf mark howison e wes bethel prabhat pdf 354k journal paper communication requirements and interconnect optimization for high end scientific applications ieee transactions on parallel and distributed systems tpds march 2009 shoaib kamil leonid oliker ali pinar john shalf preprint pdf 8467k paper optimizing collective communication on multicores hotpar 2009 berkeley ca usa march 2009 rajesh nishtala and katherine yelick paper pdf 400kb 2008 journal paper optimization and performance modeling of stencil computations on modern microprocessors siam review december 2008 kaushik datta shoaib kamil samuel williams leonid oliker john shalf and katherine yelick pdf 2 8 mb paper benchmarking gpus to tune dense linear algebra supercomputing 2008 november 2008 winner best student paper vasily volkov and james demmel pdf 648k paper communication avoiding gaussian elimination supercomputing 2008 november 2008 laura grigori james demmel and hua xiang pdf 584k paper stencil computation optimization and auto tuning on state of the art multicore architectures supercomputing 2008 november 2008 kaushik datta mark murphy vasily volkov samuel williams jonathan carter leonid oliker david patterson john shalf and katherine yelick pdf 612k talk slides ppt 12 9 mb paper hybrid electric photonic networks for scientific applications on tiled cmps workshop on high performance embedding computing august 2008 ankit jain shoaib kamil marghoob mohiyuddin john shalf and john d kubiatowicz abstract pdf 282k talk slides ppt 5 1 mb masters report poski an extensible autotuning framework to perform optimized spmvs on multicore architectures ankit jain pdf 5 3 mb tech report lu qr and cholesky factorizations using vector capabilities of gpus ucb eecs 2008 49 may 2008 vasily volkov and james demmel pdf 1 0 mb paper avoiding communication in sparse matrix computations ieee international parallel and distributed processing symposium april 2008 james demmel mark hoemmen marghoob mohiyuddin and katherine yelick pdf 1 mb talk slides pdf 6 2 mb paper lattice boltzmann simulation optimization on leading multicore platforms ieee international parallel and distributed processing symposium april 2008 winner best paper for applications track samuel williams jonathan carter leonid oliker john shalf and katherine yelick pdf 560k talk slides pdf 10 4 mb ppt 2 6 mb 2007 tech report using gpus to accelerate the bisection algorithm for finding eigenvalues of symmetric tridiagonal matrices ucb eecs 2007 179 december 2007 vasily volkov and james demmel pdf 1 8 mb paper optimization of sparse matrix vector multiplication on emerging multicore platforms supercomputing 2007 november 2007 samuel williams leonid oliker richard vuduc john shalf katherine yelick and james demmel pdf 438k talk slides pdf 6 4 mb ppt 2 5 mb tech report avoiding communication in computing krylov subspaces ucb eecs 2007 123 october 2007 james demmel mark hoemmen marghoob mohiyuddin and katherine yelick pdf 34 9 mb journal paper scientific computing kernels on the cell processor international journal of parallel programming april 2007 samuel williams john shalf leonid oliker shoaib kamil parry husbands and katherine yelick pdf 376k journal paper when cache blocking sparse matrix vector multiply works and why applicable algebra in engineering communication and computin...