{"id":42134,"date":"2019-03-18T00:00:00","date_gmt":"2019-03-17T21:00:00","guid":{"rendered":"https:\/\/prohoster.info\/blog\/blog_prohoster\/uskorenie-fajlovogo-vvoda-vyvoda-c-c-ne-osobo-napryagayas"},"modified":"2020-02-18T13:44:56","modified_gmt":"2020-02-18T10:44:56","slug":"uskorenie-fajlovogo-vvoda-vyvoda-c-c-ne-osobo-napryagayas","status":"publish","type":"post","link":"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/uskorenie-fajlovogo-vvoda-vyvoda-c-c-ne-osobo-napryagayas","title":{"rendered":"Zrychlen\u00ed souborov\u00e9ho vstupu a v\u00fdstupu C\/C++, bez velk\u00e9ho \u00fasil\u00ed","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

\"Zrychlen\u00ed<\/p>\n

\u00davod<\/h4>\n

\nNa sv\u011bt\u011b existuje takov\u00e1 jednoduch\u00e1 a velmi u\u017eite\u010dn\u00e1 utilita \u2014 BDelta<\/a><\/noindex>, a tak se stalo, \u017ee se velmi dlouho usadila v na\u0161em v\u00fdrobn\u00edm procesu (pravda, jej\u00ed verzi se nepoda\u0159ilo nainstalovat, ale rozhodn\u011b to nebyla posledn\u00ed dostupn\u00e1). Pou\u017e\u00edv\u00e1me ji k jej\u00edmu p\u0159\u00edm\u00e9mu \u00fa\u010delu \u2014 vytv\u00e1\u0159en\u00ed bin\u00e1rn\u00edch patch\u016f. Kdy\u017e se pod\u00edv\u00e1te, co tam je v repozit\u00e1\u0159i, \u2014 je to trochu smutn\u00e9: v podstat\u011b je d\u00e1vno opu\u0161t\u011bn a mnoho z toho je siln\u011b zastaral\u00e9 (kdysi tam u\u010dinil n\u011bkolik \u00faprav m\u016fj b\u00fdval\u00fd kolega, ale to u\u017e bylo d\u00e1vno). Rozhodl jsem se toto o\u017eivit: fork, odstranil jsem to, co nepl\u00e1nuji pou\u017e\u00edvat, a p\u0159ev\u00e1d\u00edm projekt na cmake<\/a><\/noindex>, za-inlinil jsem \u201ehork\u00e9\u201c mikrofunkce, odstranil jsem z z\u00e1sobn\u00edku velk\u00e9 pole (a pole prom\u011bnn\u00e9 d\u00e9lky, kter\u00e9 m\u011b up\u0159\u00edmn\u011b \u201e\u0161tvou\u201c), znovu jsem spustil profiler \u2014 a zjistil jsem, \u017ee asi 40% \u010dasu se tr\u00e1v\u00ed na fwrite<\/a><\/noindex>\u2026
\n<\/a><\/noindex><\/p>\n

Co tam m\u00e1me s fwrite?<\/h4>\n

\nV tomto k\u00f3du se fwrite (v m\u00e9m konkr\u00e9tn\u00edm testovac\u00edm p\u0159\u00edpad\u011b: vytv\u00e1\u0159en\u00ed patchu mezi bl\u00edzk\u00fdmi 300 MB soubory, vstupy jsou zcela v pam\u011bti) vol\u00e1 milionkr\u00e1t s malou velikost\u00ed bufferu. Je z\u0159ejm\u00e9, \u017ee tahle v\u011bc bude zpomalovat, a proto bych cht\u011bl n\u011bjak ovlivnit tento nepo\u0159\u00e1dek. Nechci zav\u00e1d\u011bt r\u016fzn\u00e9 zdroje dat, asynchronn\u00ed vstup\/v\u00fdstup, cht\u011bl jsem naj\u00edt jednodu\u0161\u0161\u00ed \u0159e\u0161en\u00ed. Prvn\u00ed, co m\u011b napadlo \u2014 zv\u00fd\u0161it velikost bufferu<\/p>\n

setvbuf(file, nullptr, _IOFBF, 64 * 1024)<\/code><\/pre>\n
\nale z\u00e1sadn\u00edho zlep\u0161en\u00ed v\u00fdsledku jsem nedos\u00e1hl (nyn\u00ed na fwrite odpadalo asi 37% \u010dasu) \u2014 znamen\u00e1 to, \u017ee probl\u00e9m nen\u00ed v \u010dast\u00e9m z\u00e1pisu dat na disk. Kdy\u017e se pod\u00edv\u00e1te \u201epod kapotu\u201c fwrite, m\u016f\u017eete vid\u011bt, \u017ee uvnit\u0159 se d\u011bje lock\/unlock struktury FILE p\u0159ibli\u017en\u011b takto (pseudok\u00f3d, cel\u00e1 anal\u00fdza byla prov\u00e1d\u011bna pod Visual Studio 2017):<\/p>\n
\nsize_t fwrite (const void *buffer, size_t size, size_t count, FILE *stream)\n{\n   size_t retval = 0;\n   _lock_str(stream);   \/* lock stream *\/\n   __try\n   {\n      retval = _fwrite_nolock(buffer, size, count, stream);\n   }\n   __finally \n   {\n       _unlock_str(stream);   \/* unlock stream *\/\n   }\n   return retval;\n}\n<\/code><\/pre>\n
\nPokud v\u011b\u0159\u00edte profileru, na _fwrite_nolock p\u0159ipad\u00e1 pouze 6% \u010dasu, zbytek \u2014 na overhead. V m\u00e9m konkr\u00e9tn\u00edm p\u0159\u00edpad\u011b je vl\u00e1kno bezpe\u010dnost z\u0159ejm\u00fdm p\u0159ebytkem, a proto ji ob\u011btuji, kdy\u017e nahrad\u00edm vol\u00e1n\u00ed fwrite na _fwrite_nolock<\/a><\/noindex> \u2014 dokonce nen\u00ed t\u0159eba \u0161pekulovat s argumenty. Celkov\u011b: tato prost\u00e1 manipulace podstatn\u011b sn\u00ed\u017eila n\u00e1klady na z\u00e1znam v\u00fdsledku, kter\u00e9 v p\u016fvodn\u00ed variant\u011b p\u0159edstavovaly t\u00e9m\u011b\u0159 polovinu \u010dasov\u00fdch n\u00e1rok\u016f. Mimochodem, ve sv\u011bt\u011b POSIX existuje podobn\u00e1 funkce \u2014 fwrite_unlocked<\/a><\/noindex>. Obecn\u011b plat\u00ed, \u017ee to sam\u00e9 se t\u00fdk\u00e1 i fread. T\u00edmto zp\u016fsobem lze pomoc\u00ed p\u00e1r definic #define dos\u00e1hnout zcela multiplatformn\u00edho \u0159e\u0161en\u00ed bez zbyte\u010dn\u00e9ho blokov\u00e1n\u00ed, pokud to nen\u00ed nezbytn\u00e9 (co\u017e se velmi \u010dasto st\u00e1v\u00e1).<\/p>\n
fwrite, _fwrite_nolock, setvbuf<\/h4>\n
\nPoj\u010fme se abstrahovat od p\u016fvodn\u00edho projektu a zam\u011b\u0159me se na testov\u00e1n\u00ed konkr\u00e9tn\u00edho p\u0159\u00edpadu: z\u00e1znam velk\u00e9ho souboru (512 MB) v minim\u00e1ln\u00edch porc\u00edch \u2014 po 1 bajtu. Testovac\u00ed syst\u00e9m: AMD Ryzen 7 1700, 16 GB RAM, HDD 7200 ot.\/min 64 MB cache, Windows 10 1809, bin\u00e1rn\u00ed soubor byl sestaven jako 32bitov\u00fd, optimalizace zapnuty, knihovna staticky p\u0159ilinkov\u00e1na.<\/p>\n
Vzorek pro proveden\u00ed experimentu:<\/p>\n
\n#include <chrono>\n#include <cstdio>\n#include <inttypes.h>\n#include <memory>\n\n#ifdef _MSC_VER\n#define fwrite_unlocked _fwrite_nolock\n#endif\n\nusing namespace std::chrono;\n\nint main()\n{\n    std::unique_ptr<FILE, int(*)(FILE*)> file(fopen(\"test.bin\", \"wb\"), fclose);\n    if (!file)\n        return 1;\n\n    constexpr size_t TEST_BUFFER_SIZE = 256 * 1024;\n    if (setvbuf(file.get(), nullptr, _IOFBF, TEST_BUFFER_SIZE) != 0)\n        return 2;\n\n    auto start = steady_clock::now();\n    const uint8_t b = 77;\n    constexpr size_t TEST_FILE_SIZE = 512 * 1024 * 1024;\n    for (size_t i = 0; i < TEST_FILE_SIZE; ++i)\n        fwrite_unlocked(&b, 1, sizeof(b), file.get());\n\n    auto end = steady_clock::now();\n    auto interval = duration_cast<microseconds>(end - start);\n    printf(\"Time: %lldn\", interval.count());\n\n    return 0;\n}\n<\/code><\/pre>\n
\nJako prom\u011bnn\u00e9 budou vystupovat TEST_BUFFER_SIZE, a pro p\u00e1r p\u0159\u00edpad\u016f nahrad\u00edme fwrite_unlocked za fwrite. Za\u010dneme s p\u0159\u00edpadem fwrite bez explicitn\u00edho nastaven\u00ed velikosti buffru (zakomentujeme setvbuf a souvisej\u00edc\u00ed k\u00f3d): doba 27048906 \u03bcs, rychlost z\u00e1pisu \u2014 18.93 MB\/s. Nyn\u00ed nastav\u00edme velikost buffru na 64 KB: doba \u2014 25037111 \u03bcs, rychlost \u2014 20.44 MB\/s. Nyn\u00ed otestujeme pr\u00e1ci _fwrite_nolock bez vol\u00e1n\u00ed setvbuf: 7262221 \u03bcs, rychlost \u2014 70.5 MB\/s! <\/p>\n
Pokra\u010dujeme s experimentov\u00e1n\u00edm s velikost\u00ed buffru (setvbuf):<\/p>\n
\"Zrychlen\u00ed
\n
\nData byla z\u00edsk\u00e1na pr\u016fm\u011brov\u00e1n\u00edm 5 experiment\u016f, na chyby jsem se neobt\u011b\u017eoval po\u010d\u00edtat. Podle m\u00e9ho n\u00e1zoru je 93 MB\/s p\u0159i z\u00e1pisu po 1 bajtu na b\u011b\u017en\u00e9m HDD velmi slu\u0161n\u00fd v\u00fdsledek, sta\u010d\u00ed si jen vybrat optim\u00e1ln\u00ed velikost buffru (v m\u00e9m p\u0159\u00edpad\u011b 256 KB \u2014 to je pr\u00e1v\u011b ono) a nahradit fwrite za _fwrite_nolock\/fwrite_unlocked (pokud nen\u00ed pot\u0159eba bezpe\u010dnost pro v\u00edce vl\u00e1ken, samoz\u0159ejm\u011b). 
\nPodobn\u011b s fread za podobn\u00fdch podm\u00ednek. Proto\u017ee nem\u00e1m po ruce 'hardwarov\u00fd' stroj s linuxem (jednodeskov\u00e9 po\u010d\u00edta\u010de neberu v \u00favahu), rozhodl jsem se prov\u00e9st omezen\u00fd experiment na virtu\u00e1lce (Hyper-V, OpenSUSE 15, GCC 8.3.1) \u2014 z\u00e1konitost je v z\u00e1sad\u011b stejn\u00e1: 'hol\u00e9' fwrite 20 MB\/s, fwrite + buffr na 256 KB dos\u00e1hl 23 MB\/s, fwrite_unlocked s t\u00edmto stejn\u00fdm buffrem \u2014 35 MB\/s (bin\u00e1rn\u00ed soubor 64bitov\u00fd, kompilovalo se g++ -o2 -s -static-libgcc -static-libstdc++ fwrite_test.cpp -o fwrite_test).<\/p>\n
Pozn\u00e1mka na z\u00e1v\u011br<\/h4>\n
\nC\u00edlem tohoto \u010dl\u00e1nku bylo popsat jednoduchou a \u00fa\u010dinnou techniku, kter\u00e1 se v mnoha p\u0159\u00edpadech osv\u011bd\u010duje (s funkcemi _fwrite_nolock\/fwrite_unlocked jsem se p\u0159edt\u00edm p\u0159\u00edli\u0161 nesetkal, i kdy\u017e jsou pom\u011brn\u011b m\u00e1lo zn\u00e1m\u00e9 \u2014 co\u017e je \u0161koda). Na originalitu materi\u00e1lu si nehraji, ale douf\u00e1m, \u017ee \u010dl\u00e1nek bude pro komunitu u\u017eite\u010dn\u00fd.<\/p>\n
Zdroj: habr.com<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"
\u041f\u0440\u0435\u0434\u0438\u0441\u043b\u043e\u0432\u0438\u0435 \u0415\u0441\u0442\u044c \u043d\u0430 \u0441\u0432\u0435\u0442\u0435 \u0442\u0430\u043a\u0430\u044f \u043f\u0440\u043e\u0441\u0442\u0430\u044f \u0438 \u043e\u0447\u0435\u043d\u044c \u043f\u043e\u043b\u0435\u0437\u043d\u0430\u044f \u0443\u0442\u0438\u043b\u0438\u0442\u0430 \u2014 BDelta, \u0438 \u0442\u0430\u043a \u0432\u044b\u0448\u043b\u043e, \u0447\u0442\u043e \u043e\u043d\u0430 \u043e\u0447\u0435\u043d\u044c \u0434\u0430\u0432\u043d\u043e \u0443\u043a\u043e\u0440\u0435\u043d\u0438\u043b\u0430\u0441\u044c \u0432 \u043d\u0430\u0448\u0435\u043c \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u043c \u043f\u0440\u043e\u0446\u0435\u0441\u0441\u0435 (\u043f\u0440\u0430\u0432\u0434\u0430 \u0435\u0451 \u0432\u0435\u0440\u0441\u0438\u044e \u0443\u0441\u0442\u0430\u043d\u043e\u0432\u0438\u0442\u044c \u043d\u0435 \u0443\u0434\u0430\u043b\u043e\u0441\u044c, \u043d\u043e \u043e\u043d\u0430 \u0442\u043e\u0447\u043d\u043e \u0431\u044b\u043b\u0430 \u043d\u0435 \u043f\u043e\u0441\u043b\u0435\u0434\u043d\u0435\u0439 \u0434\u043e\u0441\u0442\u0443\u043f\u043d\u043e\u0439). \u0418\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u0435\u043c \u0435\u0451 \u043f\u043e \u043f\u0440\u044f\u043c\u043e\u043c\u0443 \u043d\u0430\u0437\u043d\u0430\u0447\u0435\u043d\u0438\u044e \u2014 \u043f\u043e\u0441\u0442\u0440\u043e\u0435\u043d\u0438\u0435 \u0431\u0438\u043d\u0430\u0440\u043d\u044b\u0445 \u043f\u0430\u0442\u0447\u0435\u0439. \u0415\u0441\u043b\u0438 \u0432\u0437\u0433\u043b\u044f\u043d\u0443\u0442\u044c, \u0447\u0442\u043e \u0442\u0430\u043c \u0432 \u0440\u0435\u043f\u043e\u0437\u0438\u0442\u043e\u0440\u0438\u0438, \u2014 \u0441\u0442\u0430\u043d\u043e\u0432\u0438\u0442\u0441\u044f \u0441\u043b\u0435\u0433\u043a\u0430 […]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[688],"tags":[],"class_list":["post-42134","post","type-post","status-publish","format-standard","hentry","category-administrirovanie"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\n","aioseo_head_json":{"title":"\ud83e\udd47Zrychlen\u00ed souborov\u00e9ho vstupu a v\u00fdstupu C\/C++, bez velk\u00e9ho \u00fasil\u00ed | ProHoster","description":"P\u0159edmluva.","canonical_url":"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/uskorenie-fajlovogo-vvoda-vyvoda-c-c-ne-osobo-napryagayas","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"cs_CZ","og:site_name":"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b","og:type":"article","og:title":"\ud83e\udd47\u0423\u0441\u043a\u043e\u0440\u0435\u043d\u0438\u0435 \u0444\u0430\u0439\u043b\u043e\u0432\u043e\u0433\u043e \u0432\u0432\u043e\u0434\u0430-\u0432\u044b\u0432\u043e\u0434\u0430 C\/C++, \u043d\u0435 \u043e\u0441\u043e\u0431\u043e \u043d\u0430\u043f\u0440\u044f\u0433\u0430\u044f\u0441\u044c | ProHoster","og:description":"\u041f\u0440\u0435\u0434\u0438\u0441\u043b\u043e\u0432\u0438\u0435 \u0415\u0441\u0442\u044c.","og:url":"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/uskorenie-fajlovogo-vvoda-vyvoda-c-c-ne-osobo-napryagayas","og:image":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:secure_url":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:width":350,"og:image:height":350,"article:published_time":"2019-03-17T21:00:00+00:00","article:modified_time":"2020-02-18T10:44:56+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"42134","title":null,"description":null,"keywords":null,"keyphrases":null,"primary_term":null,"canonical_url":null,"og_title":null,"og_description":null,"og_object_type":"default","og_image_type":"default","og_image_url":null,"og_image_width":null,"og_image_height":null,"og_image_custom_url":null,"og_image_custom_fields":null,"og_video":null,"og_custom_url":null,"og_article_section":null,"og_article_tags":null,"twitter_use_og":false,"twitter_card":"default","twitter_image_type":"default","twitter_image_url":null,"twitter_image_custom_url":null,"twitter_image_custom_fields":null,"twitter_title":null,"twitter_description":null,"schema":{"blockGraphs":[],"customGraphs":[],"default":{"data":{"Article":[],"Course":[],"Dataset":[],"FAQPage":[],"Movie":[],"Person":[],"Product":[],"ProductReview":[],"Car":[],"Recipe":[],"Service":[],"SoftwareApplication":[],"WebPage":[]},"graphName":"","isEnabled":true},"graphs":[]},"schema_type":null,"schema_type_options":null,"pillar_content":false,"robots_default":true,"robots_noindex":false,"robots_noarchive":false,"robots_nosnippet":false,"robots_nofollow":false,"robots_noimageindex":false,"robots_noodp":false,"robots_notranslate":false,"robots_max_snippet":null,"robots_max_videopreview":null,"robots_max_imagepreview":"large","priority":null,"frequency":null,"local_seo":null,"seo_analyzer_scan_date":"2026-01-22 07:22:19","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2021-03-01 00:05:22","updated":"2026-01-22 07:22:19","focus_keyword":null,"additional_keywords":null,"truseo_locale":null},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts\/42134","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/comments?post=42134"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts\/42134\/revisions"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/media?parent=42134"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/categories?post=42134"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/tags?post=42134"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}