Поиск по:classic.chem.msu.su -
Поискать по всем серверам
На этой странице приведены все страницы сервера classic.chem.msu.su ,которые мы индексируем. Показаны документы 161 - 165 из 165.
Упорядочить по:
URL
|
дате изменения
161. http://classic.chem.msu.su/gran/gamess/tutor.pdf
... How GAMESS is used in chemical research? ... Variety of algorithms and data structures. ... Source-level changes. ... Development of our own QC code. ... Original code sequence: DO MB=1,NVIR DO MJ=1,NOC DO MK=1,NOC DO MA=1,NVIR MAI = IA(MA+NOC) TERM = T(MA,MJ,MK) DO MI=1,NOC MAI = MAI+1 DIKAB = E(MI) + E(MK) - E(MA+NOC) - E(MB+NOC) P(MI,MJ) = P(MI,MJ) - TERM*X(MK,MAI,MB+NOC)/DIKAB END DO END DO END DO END DO END DO Use of optimized libraries: Example. of optimized libraries: Example. ...
[
Текст
]
Ссылки http://classic.chem.msu.su/gran/gamess/tutor.pdf -- 281.5 Кб -- 25.06.2006
Похожие документы
Похожие документы
162. http://classic.chem.msu.su/gran/gamess/bof.pdf
... The program is strongly modified to achieve the maximum possible performance on Intel-based platforms Functionally extended to those QC methods which are not currently present in the regular GAMESS version Written to support both shared memory (via multithreading on SMP systems) and distributed memory (via MPI on LANs and PC clusters) parallel models of execution Operational on all popular PC Operating Systems: Win32: NT (the preferred OS for PC ... SMP scaling vs. number of CPUs used. ...
[
Текст
]
Ссылки http://classic.chem.msu.su/gran/gamess/bof.pdf -- 105.4 Кб -- 25.06.2006
Похожие документы
Похожие документы
163. http://classic.chem.msu.su/gran/gamess/qdpt2.pdf
... Straightforward implementation of the summation of the PT series is very inefficient on modern computer architectures because: At least one (or more) slow and typically not pipelined divide operation is required to calculate each individual term of the PT series Summation runs over large amount of data involving some combinations of transformed two-electron integrals, so that it is typically not processor cache-friendly Our goals Address both these problems: 1. ...
[
Текст
]
Ссылки http://classic.chem.msu.su/gran/gamess/qdpt2.pdf -- 156.1 Кб -- 25.06.2006
Похожие документы
Похожие документы
164. http://classic.chem.msu.su/gran/gamess/mp2par.pdf
... Parallelization problems Parallel code goals On the first pass, different qs distributed over nodes. ... Composite ij index Parallel implicit sorting, worker thread view Outgoing messages buffers in memory Node 1 Node 2 Node 3 Node 4 Composite ij index Parallel implicit sorting, P2P thread view P2P incoming messages Buffers (slices) in memory Node 1 Node 2 Node 3 Node 4 Lists of records Direct access file records How it works: Second pass : presorted half-transformed ...
[
Текст
]
Ссылки http://classic.chem.msu.su/gran/gamess/mp2par.pdf -- 303.0 Кб -- 06.06.2006
Похожие документы
Похожие документы
165. http://classic.chem.msu.su/gran/gamess/guideline.doc
Guideline for building and setting-up a win32x64 PC computing system running NT-family Microsoft OS with PC GAMESS 6.x. by Serge V. Kovrigin, IPMS & Kiev National University, ©2005 Table of Contents: 1. ... It is foolish to purchase high-end expensive processor and one-HDD storage-system (even if it is a SCSI disk), as more likely than work fast such system will work slower compared to not so fast processor and a 2-disk RAID (see below). ... Periodically check you PC GAMESS disk (RAID) for errors. ...
[
Текст
]
Ссылки http://classic.chem.msu.su/gran/gamess/guideline.doc -- 76.0 Кб -- 11.05.2006
Похожие документы
Похожие документы