Matches in SemOpenAlex for { <https://semopenalex.org/work/W2407804532> ?p ?o ?g. }
- W2407804532 endingPage "160" @default.
- W2407804532 startingPage "1" @default.
- W2407804532 abstract "Many crucial scientific computing applications, ranging from national security to medical advances, rely on high-performance linear algebra algorithms and technologies, underscoring their importance and broad impact. Here we present the state-of-the-art design and implementation practices for the acceleration of the predominant linear algebra algorithms on large-scale accelerated multicore systems. Examples are given with fundamental dense linear algebra algorithms – from the LU, QR, Cholesky, and LDLT factorizations needed for solving linear systems of equations, to eigenvalue and singular value decomposition (SVD) problems. The implementations presented are readily available via the open-source PLASMA and MAGMA libraries, which represent the next generation modernization of the popular LAPACK library for accelerated multicore systems. To generate the extreme level of parallelism needed for the efficient use of these systems, algorithms of interest are redesigned and then split into well-chosen computational tasks. The task execution is scheduled over the computational components of a hybrid system of multicore CPUs with GPU accelerators and/or Xeon Phi coprocessors, using either static scheduling or light-weight runtime systems. The use of light-weight runtime systems keeps scheduling overheads low, similar to static scheduling, while enabling the expression of parallelism through sequential-like code. This simplifies the development effort and allows exploration of the unique strengths of the various hardware components. Finally, we emphasize the development of innovative linear algebra algorithms using three technologies – mixed precision arithmetic, batched operations, and asynchronous iterations – that are currently of high interest for accelerated multicore systems." @default.
- W2407804532 created "2016-06-24" @default.
- W2407804532 creator A5012133869 @default.
- W2407804532 creator A5023168533 @default.
- W2407804532 creator A5052397462 @default.
- W2407804532 creator A5070456582 @default.
- W2407804532 creator A5073990539 @default.
- W2407804532 creator A5075517045 @default.
- W2407804532 creator A5077268543 @default.
- W2407804532 creator A5082630993 @default.
- W2407804532 creator A5083604741 @default.
- W2407804532 creator A5087711762 @default.
- W2407804532 date "2016-05-01" @default.
- W2407804532 modified "2023-10-14" @default.
- W2407804532 title "Linear algebra software for large-scale accelerated multicore computing" @default.
- W2407804532 cites W106544777 @default.
- W2407804532 cites W133977063 @default.
- W2407804532 cites W1481082265 @default.
- W2407804532 cites W1490374280 @default.
- W2407804532 cites W1505489746 @default.
- W2407804532 cites W1506342804 @default.
- W2407804532 cites W1558296484 @default.
- W2407804532 cites W1569981719 @default.
- W2407804532 cites W1575701986 @default.
- W2407804532 cites W1581582143 @default.
- W2407804532 cites W1589630408 @default.
- W2407804532 cites W1737435842 @default.
- W2407804532 cites W177735991 @default.
- W2407804532 cites W1802445103 @default.
- W2407804532 cites W1973166496 @default.
- W2407804532 cites W1973702207 @default.
- W2407804532 cites W1975721804 @default.
- W2407804532 cites W1975818677 @default.
- W2407804532 cites W1976964180 @default.
- W2407804532 cites W1977120413 @default.
- W2407804532 cites W1977146902 @default.
- W2407804532 cites W1981367360 @default.
- W2407804532 cites W1981452262 @default.
- W2407804532 cites W1988201922 @default.
- W2407804532 cites W1988425770 @default.
- W2407804532 cites W1991973947 @default.
- W2407804532 cites W1996275648 @default.
- W2407804532 cites W2002257715 @default.
- W2407804532 cites W2006742127 @default.
- W2407804532 cites W2007768252 @default.
- W2407804532 cites W2008946067 @default.
- W2407804532 cites W2011777324 @default.
- W2407804532 cites W2015806545 @default.
- W2407804532 cites W2016279572 @default.
- W2407804532 cites W2016358047 @default.
- W2407804532 cites W2020804487 @default.
- W2407804532 cites W2022842990 @default.
- W2407804532 cites W2026237859 @default.
- W2407804532 cites W2031392034 @default.
- W2407804532 cites W2032576830 @default.
- W2407804532 cites W2036603155 @default.
- W2407804532 cites W2037798332 @default.
- W2407804532 cites W2038469228 @default.
- W2407804532 cites W2039778940 @default.
- W2407804532 cites W2041876368 @default.
- W2407804532 cites W2042879774 @default.
- W2407804532 cites W2045618500 @default.
- W2407804532 cites W2048255281 @default.
- W2407804532 cites W2048372372 @default.
- W2407804532 cites W2049790251 @default.
- W2407804532 cites W2052926498 @default.
- W2407804532 cites W2053489736 @default.
- W2407804532 cites W2057081399 @default.
- W2407804532 cites W2059900236 @default.
- W2407804532 cites W2060236641 @default.
- W2407804532 cites W2060803192 @default.
- W2407804532 cites W2063675347 @default.
- W2407804532 cites W2064733273 @default.
- W2407804532 cites W2065402594 @default.
- W2407804532 cites W2067580655 @default.
- W2407804532 cites W2070299075 @default.
- W2407804532 cites W2070525241 @default.
- W2407804532 cites W2070571481 @default.
- W2407804532 cites W2071128523 @default.
- W2407804532 cites W2073126584 @default.
- W2407804532 cites W2073260424 @default.
- W2407804532 cites W2079120751 @default.
- W2407804532 cites W2079461498 @default.
- W2407804532 cites W2079925720 @default.
- W2407804532 cites W2080090223 @default.
- W2407804532 cites W2080710904 @default.
- W2407804532 cites W2083715026 @default.
- W2407804532 cites W2085850802 @default.
- W2407804532 cites W2091487980 @default.
- W2407804532 cites W2095258827 @default.
- W2407804532 cites W2095420020 @default.
- W2407804532 cites W2100530229 @default.
- W2407804532 cites W2102181290 @default.
- W2407804532 cites W2107857731 @default.
- W2407804532 cites W2111593426 @default.
- W2407804532 cites W2113830538 @default.
- W2407804532 cites W2115780241 @default.
- W2407804532 cites W2117782600 @default.