{"id":34335,"date":"2019-10-31T21:57:43","date_gmt":"2019-10-31T18:57:43","guid":{"rendered":"https:\/\/prohoster.info\/blog\/analiz-proizvoditelnosti-virtualnoj-mashiny-v-vmware-vsphere-chast-1-cpu\/"},"modified":"2019-10-31T21:57:43","modified_gmt":"2019-10-31T18:57:43","slug":"analiz-proizvoditelnosti-virtualnoj-mashiny-v-vmware-vsphere-chast-1-cpu","status":"publish","type":"post","link":"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/analiz-proizvoditelnosti-virtualnoj-mashiny-v-vmware-vsphere-chast-1-cpu","title":{"rendered":"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/0c077a1754c2e32c1f339999ceacf03a.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nPokud spravujete virtu\u00e1ln\u00ed infrastrukturu na b\u00e1zi VMware vSphere (nebo jak\u00e9hokoli jin\u00e9ho technologick\u00e9ho stacku), pravd\u011bpodobn\u011b \u010dasto sl\u00fdch\u00e1te od u\u017eivatel\u016f st\u00ed\u017enosti: \u201eVirtu\u00e1ln\u00ed stroj b\u011b\u017e\u00ed pomalu!\u201c. V t\u00e9to s\u00e9rii \u010dl\u00e1nk\u016f si podrobn\u011b probereme metriky v\u00fdkonnosti a pov\u00edme si, co a pro\u010d \u201ezpomaluj\u00ed\u201c a jak tomu p\u0159edej\u00edt.<\/p>\n<p>Budu zva\u017eovat n\u00e1sleduj\u00edc\u00ed aspekty v\u00fdkonnosti virtu\u00e1ln\u00edch stroj\u016f:<\/p>\n<ul>\n<li>CPU,<\/li>\n<li>RAM,<\/li>\n<li>DISK,<\/li>\n<li>S\u00ed\u0165.<\/li>\n<\/ul>\n<p>\nZa\u010dnu s CPU.<\/p>\n<p>K anal\u00fdze v\u00fdkonnosti budeme pot\u0159ebovat:<\/p>\n<ul>\n<li><b>vCenter Performance Counters<\/b> \u2013 v\u00fdkonov\u00e9 ukazatele, jejich\u017e grafy si m\u016f\u017eete prohl\u00e9dnout p\u0159es vSphere Client. Informace o t\u011bchto ukazatel\u00edch jsou dostupn\u00e9 v jak\u00e9koli verzi klienta (\"tlust\u00fd\" klient na C#, webov\u00fd klient na Flex a webov\u00fd klient na HTML5). V t\u011bchto \u010dl\u00e1nc\u00edch budeme pou\u017e\u00edvat screenshoty z C# klienta, pouze proto, \u017ee vypadaj\u00ed l\u00e9pe v miniaturn\u00ed podob\u011b :)<\/li>\n<li><b>ESXTOP<\/b> \u2013 n\u00e1stroj, kter\u00fd lze spustit z p\u0159\u00edkazov\u00e9ho \u0159\u00e1dku ESXi. Pomoc\u00ed n\u011bj m\u016f\u017eete z\u00edskat hodnoty v\u00fdkonov\u00fdch ukazatel\u016f v re\u00e1ln\u00e9m \u010dase nebo exportovat tyto hodnoty za ur\u010dit\u00e9 obdob\u00ed do .csv souboru pro dal\u0161\u00ed anal\u00fdzu. Pozd\u011bji se o tomto n\u00e1stroji zm\u00edn\u00edm podrobn\u011bji a p\u0159id\u00e1m n\u011bkolik u\u017eite\u010dn\u00fdch odkaz\u016f na dokumentaci a \u010dl\u00e1nky na toto t\u00e9ma.<\/li>\n<\/ul>\n<p>\n<noindex><a rel=\"nofollow\" name=\"habracut\"><\/a><\/noindex><\/p>\n<h3>Trochu teorie<\/h3>\n<p>\n<img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/bc415aace7653c45850d26b7104483c9.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nV ESXi zodpov\u00edd\u00e1 za pr\u00e1ci ka\u017ed\u00e9ho vCPU (j\u00e1dra virtu\u00e1ln\u00edho stroje) samostatn\u00fd proces \u2013 world v terminologii VMware. Existuj\u00ed tak\u00e9 servisn\u00ed procesy, ale z hlediska anal\u00fdzy v\u00fdkonu VM jsou m\u00e9n\u011b zaj\u00edmav\u00e9.<\/p>\n<p>Proces v ESXi m\u016f\u017ee b\u00fdt v jednom ze \u010dty\u0159 stav\u016f:<\/p>\n<ul>\n<li><b>Spustit<\/b> \u2013 proces vykon\u00e1v\u00e1 n\u011bjakou u\u017eite\u010dnou pr\u00e1ci.<\/li>\n<li><b>\u010cek\u00e1n\u00ed<\/b> \u2013 proces nevykon\u00e1v\u00e1 \u017e\u00e1dnou pr\u00e1ci (idle) nebo \u010dek\u00e1 na vstup\/v\u00fdstup.<\/li>\n<li><b>Costop<\/b> \u2013 stav, kter\u00fd vznik\u00e1 v v\u00edcej\u00e1drov\u00fdch virtu\u00e1ln\u00edch stroj\u00edch. Doch\u00e1z\u00ed k n\u011bmu, kdy\u017e pl\u00e1nova\u010d CPU hypervizoru (ESXi CPU Scheduler) nedok\u00e1\u017ee napl\u00e1novat sou\u010dasn\u00e9 prov\u00e1d\u011bn\u00ed na fyzick\u00fdch yahdrech serveru pro v\u0161echna aktivn\u00ed j\u00e1dra virtu\u00e1ln\u00edho stroje. Ve fyzick\u00e9m sv\u011bt\u011b pracuj\u00ed v\u0161echna j\u00e1dra procesoru paraleln\u011b, hostitelsk\u00fd opera\u010dn\u00ed syst\u00e9m uvnit\u0159 VM po\u010d\u00edt\u00e1 s podobn\u00fdm chov\u00e1n\u00edm, a proto je hypervizor nucen zpomalit j\u00e1dra VM, kter\u00e1 maj\u00ed mo\u017enost dokon\u010dit cyklus rychleji. V modern\u00edch verz\u00edch ESXi pl\u00e1nova\u010d CPU pou\u017e\u00edv\u00e1 mechanismus, kter\u00fd se naz\u00fdv\u00e1 relaxed co-scheduling: hypervizor posuzuje rozd\u00edl mezi nejrychlej\u0161\u00edm a nejpomalej\u0161\u00edm j\u00e1drem virtu\u00e1ln\u00edho stroje (skew). Pokud rozd\u00edl p\u0159ekro\u010d\u00ed stanovenou mez, \u201erychl\u00e9\u201c j\u00e1dro p\u0159ech\u00e1z\u00ed do stavu costop. Pokud j\u00e1dra VM tr\u00e1v\u00ed hodn\u011b \u010dasu v tomto stavu, m\u016f\u017ee to zp\u016fsobit probl\u00e9my s v\u00fdkonem.<\/li>\n<li><b>Ready<\/b> \u2013 proces p\u0159ech\u00e1z\u00ed do tohoto stavu, kdy\u017e hypervizor nem\u00e1 mo\u017enost alokovat prost\u0159edky pro jeho vykon\u00e1n\u00ed. Vysok\u00e9 hodnoty ready mohou zp\u016fsobit probl\u00e9my s v\u00fdkonem VM.<\/li>\n<\/ul>\n<p><\/p>\n<h3>Hlavn\u00ed ukazatele v\u00fdkonu CPU virtu\u00e1ln\u00edho stroje<\/h3>\n<p>\n<b>Vyu\u017eit\u00ed CPU, %.<\/b> Ukazuje procento vyu\u017eit\u00ed CPU za zadan\u00e9 obdob\u00ed.<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/7a5da7dcd45329c3aae8e394c5d35a30.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\n<b>Jak analyzovat?<\/b> Pokud VM stabiln\u011b vyu\u017e\u00edv\u00e1 CPU na 90 % nebo jsou n\u00e1razy a\u017e na 100 %, m\u00e1me probl\u00e9m. Probl\u00e9my se mohou projevovat nejen v \u201epomal\u00e9\u201c pr\u00e1ci aplikace uvnit\u0159 VM, ale i v nedostupnosti VM p\u0159es s\u00ed\u0165. Pokud monitorovac\u00ed syst\u00e9m ukazuje, \u017ee VM se ob\u010das odpojuje, v\u011bnujte pozornost vrchol\u016fm v grafu vyu\u017eit\u00ed CPU.<\/p>\n<p>Existuje standardn\u00ed Alarm, kter\u00fd ukazuje zat\u00ed\u017een\u00ed CPU virtu\u00e1ln\u00edho stroje:<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/9bfd9bb5b43b9b02e8bbfac307d6c9f9.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\n<b>Co d\u011blat?<\/b> Pokud m\u00e1 VM neust\u00e1le vysok\u00e9 vyu\u017eit\u00ed CPU, je mo\u017en\u00e9 uva\u017eovat o zv\u00fd\u0161en\u00ed po\u010dtu vCPU (bohu\u017eel to nen\u00ed v\u017edy \u00fa\u010dinn\u00e9) nebo p\u0159esunout VM na server s v\u00fdkonn\u011bj\u0161\u00edmi procesory.<\/p>\n<h3>Vyu\u017eit\u00ed CPU v Mhz<\/h3>\n<p>\nV grafech na vCenter lze vid\u011bt vyu\u017eit\u00ed v % pouze cel\u00e9ho virtu\u00e1ln\u00edho stroje, grafy pro jednotliv\u00e1 j\u00e1dra nejsou (v Esxtopu jsou hodnoty % podle jader). Pro ka\u017ed\u00e9 j\u00e1dro lze vid\u011bt vyu\u017eit\u00ed v MHz.<\/p>\n<p><b>Jak analyzovat?<\/b> N\u011bkdy m\u016f\u017ee aplikace b\u00fdt neoptimalizov\u00e1na pro v\u00edcej\u00e1drovou architekturu: pou\u017e\u00edv\u00e1 na 100 % pouze jedno j\u00e1dro, zat\u00edmco ostatn\u00ed z\u016fst\u00e1vaj\u00ed bez z\u00e1t\u011b\u017ee. Nap\u0159\u00edklad p\u0159i v\u00fdchoz\u00edch nastaven\u00edch z\u00e1lohy MS SQL spou\u0161t\u00ed proces pouze na jednom j\u00e1d\u0159e. V\u00fdsledkem je, \u017ee z\u00e1lohov\u00e1n\u00ed zpomaluje ne kv\u016fli pomal\u00e9 rychlosti disk\u016f (na co si u\u017eivatel st\u011b\u017eoval), ale kv\u016fli tomu, \u017ee procesor nezvl\u00e1d\u00e1. Probl\u00e9m byl vy\u0159e\u0161en zm\u011bnou parametr\u016f: z\u00e1lohov\u00e1n\u00ed za\u010dalo b\u011b\u017eet paraleln\u011b v n\u011bkolika souborech (a t\u00edm p\u00e1dem v n\u011bkolika procesech).<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/d1bce5fe5ef89a69c31a6a8f043d5d24.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<i>P\u0159\u00edklad nevyv\u00e1\u017een\u00e9ho zat\u00ed\u017een\u00ed jader.<\/i><\/p>\n<p>M\u016f\u017ee tak\u00e9 nastat situace (jak je zn\u00e1zorn\u011bno na grafu v\u00fd\u0161e), kdy jsou j\u00e1dra zat\u00ed\u017eena nevyv\u00e1\u017een\u011b a n\u011bkter\u00e1 z nich maj\u00ed \u0161pi\u010dky na 100 %. Stejn\u011b jako p\u0159i zat\u00ed\u017een\u00ed pouze jednoho j\u00e1dra alarm na CPU Usage neobstoj\u00ed (je na cel\u00e9 VM), ale probl\u00e9my s v\u00fdkonem se projev\u00ed.<\/p>\n<p><b>Co d\u011blat? <\/b>Pokud software ve virtu\u00e1ln\u00ed ma\u0161in\u011b zat\u011b\u017euje j\u00e1dra nevyv\u00e1\u017een\u011b (pou\u017e\u00edv\u00e1 pouze jedno j\u00e1dro nebo \u010d\u00e1st jader), nen\u00ed smysl zvy\u0161ovat jejich po\u010det. V takov\u00e9m p\u0159\u00edpad\u011b je lep\u0161\u00ed p\u0159esunout VM na server s v\u00fdkonn\u011bj\u0161\u00edmi procesory.<\/p>\n<p>M\u016f\u017eete tak\u00e9 zkusit zkontrolovat nastaven\u00ed spot\u0159eby energie v BIOSu serveru. Mnoho administr\u00e1tor\u016f zap\u00edn\u00e1 v BIOSu re\u017eim Vysok\u00fd v\u00fdkon a t\u00edmto zp\u016fsobem vyp\u00ednaj\u00ed technologie \u00faspory energie C-states a P-states. V modern\u00edch procesorech Intel se pou\u017e\u00edv\u00e1 technologie Turbo Boost, kter\u00e1 zvy\u0161uje frekvenci jednotliv\u00fdch jader procesoru na \u00fakor jin\u00fdch jader. Ale funguje jen p\u0159i zapnut\u00fdch technologi\u00edch \u00faspory energie. Pokud je vypneme, procesor nem\u016f\u017ee sn\u00ed\u017eit spot\u0159ebu energie jader, kter\u00e1 nejsou zat\u00ed\u017eena. <\/p>\n<p>VMware doporu\u010duje nevyp\u00ednat technologie \u00faspory energie na serverech, ale vyb\u00edrat re\u017eimy, kter\u00e9 maxim\u00e1ln\u011b p\u0159ed\u00e1vaj\u00ed kontrolu nad spot\u0159ebou energie hypervisoru. P\u0159i tom by se v nastaven\u00edch spot\u0159eby energie hypervisoru m\u011bl vybrat re\u017eim Vysok\u00fd v\u00fdkon. <\/p>\n<p>Pokud m\u00e1te ve sv\u00e9 infrastruktu\u0159e samostatn\u00e9 VM (nebo j\u00e1dra VM), kter\u00e9 vy\u017eaduj\u00ed zv\u00fd\u0161enou frekvenci CPU, spr\u00e1vn\u00e9 nastaven\u00ed spot\u0159eby energie m\u016f\u017ee v\u00fdrazn\u011b zlep\u0161it jejich v\u00fdkon.<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/59aec2b77fa87e413d90371a11388629.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<\/p>\n<h3>CPU Ready (P\u0159ipravenost) <\/h3>\n<p>\nPokud je j\u00e1dro VM (vCPU) ve stavu Ready, neprov\u00e1d\u00ed u\u017eite\u010dnou pr\u00e1ci. Tento stav nast\u00e1v\u00e1, kdy\u017e hypervizor nenach\u00e1z\u00ed dostupn\u00e9 fyzick\u00e9 j\u00e1dro, na kter\u00e9 by mohl p\u0159i\u0159adit proces vCPU virtu\u00e1ln\u00ed stroje.<\/p>\n<p><b>Jak analyzovat?<\/b> Obvykle, pokud jsou j\u00e1dra virtu\u00e1ln\u00edho stroje ve stavu Ready v\u00edce ne\u017e 10% \u010dasu, zaznamen\u00e1te probl\u00e9my s v\u00fdkonem. Jin\u00fdmi slovy, v\u00edce ne\u017e 10% \u010dasu VM \u010dek\u00e1 na dostupnost fyzick\u00fdch zdroj\u016f.<\/p>\n<p>Ve vCenter lze sledovat 2 ukazatele souvisej\u00edc\u00ed s CPU Ready:<\/p>\n<ul>\n<li>Readiness,<\/li>\n<li>Ready.<\/li>\n<\/ul>\n<p>\nHodnoty obou ukazatel\u016f lze sledovat jak pro cel\u00fd VM, tak pro jednotliv\u00e1 j\u00e1dra.<br \/>\nReadiness ukazuje hodnotu ihned v procentech, ale pouze v re\u00e1ln\u00e9m \u010dase (data za posledn\u00ed hodinu, interval m\u011b\u0159en\u00ed 20 sekund). Tento ukazatel je lep\u0161\u00ed pou\u017e\u00edvat pouze pro identifikaci probl\u00e9m\u016f \u201ena m\u00edst\u011b\u201c.<\/p>\n<p>Hodnoty ukazatele Ready lze tak\u00e9 sledovat v historick\u00e9m kontextu. To je u\u017eite\u010dn\u00e9 pro ur\u010den\u00ed vzorc\u016f a pro hlub\u0161\u00ed anal\u00fdzu probl\u00e9mu. Nap\u0159\u00edklad, pokud m\u00e1 virtu\u00e1ln\u00ed stroj probl\u00e9my s v\u00fdkonem v ur\u010ditou dobu, lze spojit intervaly zv\u00fd\u0161en\u00e9 hodnoty CPU Ready s celkov\u00fdm zat\u00ed\u017een\u00edm serveru, na kter\u00e9m dan\u00fd VM b\u011b\u017e\u00ed, a p\u0159ijmout opat\u0159en\u00ed ke sn\u00ed\u017een\u00ed zat\u00ed\u017een\u00ed (pokud si DRS neporadilo).<\/p>\n<p>Ready na rozd\u00edl od Readiness nen\u00ed zobrazen v procentech, ale v milisekund\u00e1ch. Je to ukazatel typu Sumarizace, co\u017e znamen\u00e1, \u017ee ukazuje, jak dlouho v pr\u016fb\u011bhu m\u011b\u0159en\u00ed bylo j\u00e1dro VM ve stavu Ready. Tento \u00fadaj lze p\u0159ev\u00e9st na procenta pomoc\u00ed jednoduch\u00e9ho vzorce:<\/p>\n<p>(sou\u010det CPU ready hodnoty \/ (v\u00fdchoz\u00ed interval aktualizace grafu v sekund\u00e1ch * 1000)) * 100 = CPU ready %<\/p>\n<p>Nap\u0159\u00edklad pro VM na n\u00ed\u017ee uveden\u00e9m grafu dos\u00e1hne maxim\u00e1ln\u00ed hodnota Ready pro cel\u00fd virtu\u00e1ln\u00ed stroj n\u00e1sleduj\u00edc\u00edho: <\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/3e2df8bca55b572502d88b838cbc306e.jpg\" style=\"display:block;margin: 0 auto;\" \/> <\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/f0371c9cd655c1e4785e90debde61dc8.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nP\u0159i v\u00fdpo\u010dtu hodnoty Ready v procentech je t\u0159eba v\u011bnovat pozornost dv\u011bma bod\u016fm:<\/p>\n<ul>\n<li>Celkov\u00e9 hodnoty Ready pro VM jsou sou\u010dtem Ready pro j\u00e1dra.<\/li>\n<li>Interval m\u011b\u0159en\u00ed. Pro re\u00e1ln\u00fd \u010das \u2013 to je 20 sekund, a nap\u0159\u00edklad na denn\u00edch grafech \u2013 to je 300 sekund.<\/li>\n<\/ul>\n<p>\nP\u0159i aktivn\u00edm \u0159e\u0161en\u00ed probl\u00e9m\u016f lze tyto jednoduch\u00e9 body snadno opomenout a ztratit cenn\u00fd \u010das na \u0159e\u0161en\u00ed neexistuj\u00edc\u00edch probl\u00e9m\u016f. <\/p>\n<p>Vypo\u010d\u00edt\u00e1me Ready na z\u00e1klad\u011b \u00fadaj\u016f z grafu n\u00ed\u017ee. (324474\/(20*1000))*100 = 1622% na celou VM. Pokud se pod\u00edv\u00e1me na j\u00e1dra, v\u00fdsledek u\u017e nen\u00ed tak dramatick\u00fd: 1622\/64 = 25% na j\u00e1dro. V tomto p\u0159\u00edpad\u011b je celkem snadn\u00e9 odhalit podvod: hodnota Ready nen\u00ed realistick\u00e1. Pokud v\u0161ak hovo\u0159\u00edme o 10\u201320% na celou VM s n\u011bkolika j\u00e1dry, m\u016f\u017ee b\u00fdt hodnota na ka\u017ed\u00e9m j\u00e1dru v norm\u00e1ln\u00edch mez\u00edch.<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/703ab2b1f1fab8bda10aad781c2cef78.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\n<b>Co d\u011blat? <\/b>Vysok\u00e1 hodnota Ready nazna\u010duje, \u017ee serveru chyb\u00ed procesorov\u00e9 zdroje pro spr\u00e1vnou \u010dinnost virtu\u00e1ln\u00edch stroj\u016f. V takov\u00e9 situaci nezb\u00fdv\u00e1 ne\u017e sn\u00ed\u017eit p\u0159eklep na procesoru (vCPU:pCPU). Toho lze dos\u00e1hnout zmen\u0161en\u00edm parametr\u016f st\u00e1vaj\u00edc\u00edch VM nebo migrac\u00ed \u010d\u00e1sti VM na jin\u00e9 servery.<\/p>\n<h3>Co-stop<\/h3>\n<p>\n<b>Jak analyzovat?<\/b> Tento \u010d\u00edta\u010d m\u00e1 tak\u00e9 typ Sumace a p\u0159epo\u010d\u00edt\u00e1v\u00e1 se na procenta podobn\u011b jako Ready:<\/p>\n<p>(CPU co-stop summation value \/ (v\u00fdchoz\u00ed \u010dasov\u00fd interval aktualizace grafu v sekund\u00e1ch * 1000)) * 100 = % CPU co-stop<\/p>\n<p>Je t\u0159eba tak\u00e9 v\u011bnovat pozornost po\u010dtu jader na VM a intervalu m\u011b\u0159en\u00ed.<br \/>\nVe stavu co-stop j\u00e1dro nevykon\u00e1v\u00e1 u\u017eite\u010dnou pr\u00e1ci. P\u0159i spr\u00e1vn\u00e9m v\u00fdb\u011bru velikosti VM a norm\u00e1ln\u00edm zat\u00ed\u017een\u00ed serveru by m\u011bl \u010d\u00edta\u010d co-stop b\u00fdt bl\u00edzk\u00fd nule.<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/e226afbf91155c2d5d3fdc4571e7383a.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<i>V tomto p\u0159\u00edpad\u011b je zat\u00ed\u017een\u00ed zjevn\u011b nenorm\u00e1ln\u00ed :)<\/i><\/p>\n<p><b>Co d\u011blat?<\/b> Pokud na jednom hypervizoru b\u011b\u017e\u00ed v\u00edce VM s velk\u00fdm po\u010dtem jader a doch\u00e1z\u00ed k p\u0159eklep\u016fm na CPU, m\u016f\u017ee se \u010d\u00edta\u010d co-stop zv\u00fd\u0161it, co\u017e povede k probl\u00e9m\u016fm s v\u00fdkonem t\u011bchto VM. <\/p>\n<p>Tak\u00e9 se co-stop zv\u00fd\u0161\u00ed, pokud se pro aktivn\u00ed j\u00e1dra jedn\u00e9 VM pou\u017e\u00edvaj\u00ed vl\u00e1kna na jednom fyzick\u00e9m j\u00e1dru serveru se zapnut\u00fdm hyper-threadingem. Takov\u00e1 situace m\u016f\u017ee nastat, pokud m\u00e1 VM v\u00edce jader, ne\u017e je fyzicky dostupn\u00e9 na serveru, kde b\u011b\u017e\u00ed, nebo pokud je pro VM zapnuta volba \"preferHT\". O t\u00e9to volb\u011b si m\u016f\u017eete p\u0159e\u010d\u00edst <noindex><a rel=\"nofollow\" href=\"https:\/\/blogs.vmware.com\/vsphere\/2014\/03\/perferht-use-2.html\">zde<\/a><\/noindex>. <\/p>\n<p>Aby se p\u0159ede\u0161lo probl\u00e9m\u016fm s v\u00fdkonem VM kv\u016fli vysok\u00e9mu co-stop, vyb\u00edrejte velikost VM v souladu s doporu\u010den\u00edmi v\u00fdrobce softwaru, kter\u00fd na t\u00e9to VM b\u011b\u017e\u00ed, a podle mo\u017enost\u00ed fyzick\u00e9ho serveru, na kter\u00e9m VM pracuje. <\/p>\n<p>Ned\u00e1vejte j\u00e1dra na z\u00e1sobu, to m\u016f\u017ee zp\u016fsobit probl\u00e9my s v\u00fdkonem nejen samotn\u00e9 VM, ale i jej\u00edch soused\u016f na serveru.<\/p>\n<h3>Dal\u0161\u00ed u\u017eite\u010dn\u00e9 metriky CPU<\/h3>\n<p>\n<b>Spustit<\/b> \u2013 kolik \u010dasu (ms) v pr\u016fb\u011bhu m\u011b\u0159en\u00e9ho obdob\u00ed vCPU str\u00e1vil ve stavu RUN, tedy skute\u010dn\u011b vykon\u00e1val u\u017eite\u010dnou pr\u00e1ci.<\/p>\n<p><b>Idle<\/b> \u2013 jak dlouho (ms) byl vCPU b\u011bhem m\u011b\u0159en\u00ed v klidov\u00e9m stavu. Vysok\u00e9 hodnoty Idle nejsou probl\u00e9m, jen prost\u011b vCPU nem\u011bl \"co d\u011blat\".<\/p>\n<p><b>\u010cek\u00e1n\u00ed<\/b> \u2013 jak dlouho (ms) byl vCPU b\u011bhem m\u011b\u0159en\u00ed ve stavu Wait. Jeliko\u017e do tohoto po\u010d\u00edtadla se zahrnuje IDLE, vysok\u00e9 hodnoty Wait tak\u00e9 neznamenaj\u00ed, \u017ee by byl probl\u00e9m. Ale pokud je p\u0159i vysok\u00e9m Wait IDLE n\u00edzk\u00e9, znamen\u00e1 to, \u017ee VM \u010dekala na dokon\u010den\u00ed vstupn\u011b\/v\u00fdstupn\u00edch operac\u00ed, co\u017e m\u016f\u017ee nazna\u010dovat probl\u00e9m s v\u00fdkonem pevn\u00e9ho disku nebo n\u011bjak\u00fdch virtu\u00e1ln\u00edch za\u0159\u00edzen\u00ed VM.<\/p>\n<p><b>Maxim\u00e1ln\u00ed omezen\u00ed<\/b> \u2013 jak dlouho (ms) byl vCPU b\u011bhem m\u011b\u0159en\u00ed ve stavu Ready kv\u016fli nastaven\u00e9mu limitu na zdroje. Pokud je v\u00fdkon ne\u010dekan\u011b n\u00edzk\u00fd, je u\u017eite\u010dn\u00e9 zkontrolovat hodnotu tohoto po\u010d\u00edtadla a limit na CPU v nastaven\u00ed VM. M\u016f\u017ee se st\u00e1t, \u017ee VM m\u00e1 nastaveny limity, o kter\u00fdch nev\u00edte. Nap\u0159\u00edklad se to d\u011bje, kdy\u017e byla VM zduplikov\u00e1na z \u0161ablony, na kter\u00e9 byl nastaven limit na CPU.<\/p>\n<p><b>Swap wait<\/b> \u2013 jak dlouho vCPU \u010dekal na operace s VMkernel Swap b\u011bhem m\u011b\u0159en\u00ed. Pokud jsou hodnoty tohoto po\u010d\u00edtadla nad nulou, pak m\u00e1 VM ur\u010dit\u011b probl\u00e9my s v\u00fdkonem. O SWAPu si pov\u00edme v\u00edce v \u010dl\u00e1nku o po\u010ditadlech pam\u011bti.<\/p>\n<h3>ESXTOP<\/h3>\n<p>\nPokud jsou po\u010ditadla v\u00fdkonu ve vCenter dobr\u00e1 pro anal\u00fdzu historick\u00fdch dat, pak je lep\u0161\u00ed prov\u00e1d\u011bt opera\u010dn\u00ed anal\u00fdzu probl\u00e9mu v ESXTOP. Zde jsou v\u0161echny hodnoty prezentov\u00e1ny hotov\u011b (nen\u00ed pot\u0159eba nic p\u0159ev\u00e1d\u011bt) a minim\u00e1ln\u00ed m\u011b\u0159\u00edc\u00ed obdob\u00ed je 2 sekundy.<br \/>\nObrazovka ESXTOP pro CPU se vyvol\u00e1v\u00e1 kl\u00e1vesou \u201ec\u201c a vypad\u00e1 n\u00e1sledovn\u011b:<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/f4f3611bc383007bdbaa0948001923c4.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nPro pohodl\u00ed m\u016f\u017eete nechat pouze procesy virtu\u00e1ln\u00edch stroj\u016f stisknut\u00edm Shift-V.<br \/>\nChcete-li zobrazit metriky pro jednotliv\u00e9 j\u00e1dra VM, stiskn\u011bte \u201ee\u201c a zadejte GID z\u00e1jmov\u00e9 VM (30919 na sn\u00edmku n\u00ed\u017ee):<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/5f2c0226cee6e4e43128537bd4da843f.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nStru\u010dn\u011b projdu sloupci, kter\u00e9 jsou v\u00fdchoz\u00ed. Dal\u0161\u00ed sloupce lze p\u0159idat stisknut\u00edm \u201ef\u201c.<\/p>\n<p><b>NWLD (Po\u010det sv\u011bt\u016f)<\/b> \u2013 po\u010det proces\u016f ve skupin\u011b. Chcete-li rozbalit skupinu a zobrazit metriky pro ka\u017ed\u00fd proces (nap\u0159\u00edklad pro ka\u017ed\u00e9 j\u00e1dro v\u00edcej\u00e1drov\u00e9 VM), stiskn\u011bte \u201ce\u201d. Pokud je ve skupin\u011b v\u00edce ne\u017e jeden proces, pak jsou hodnoty metrik pro skupinu rovny sou\u010dtu metrik pro jednotliv\u00e9 procesy.<\/p>\n<p><b>%USED<\/b> \u2013 kolik cykl\u016f CPU serveru pou\u017e\u00edv\u00e1 proces nebo skupina proces\u016f.<\/p>\n<p><b>%RUN<\/b> \u2013 kolik \u010dasu byl proces ve stavu RUN, tj. vykon\u00e1val u\u017eite\u010dnou pr\u00e1ci b\u011bhem sledovan\u00e9ho obdob\u00ed. Li\u0161\u00ed se od %USED t\u00edm, \u017ee nezohled\u0148uje hyper-threading, frekven\u010dn\u00ed \u0161k\u00e1lov\u00e1n\u00ed a \u010das str\u00e1ven\u00fd na syst\u00e9mov\u00fdch \u00faloh\u00e1ch (%SYS).<\/p>\n<p><b>%SYS<\/b> \u2013 \u010das str\u00e1ven\u00fd na syst\u00e9mov\u00fdch \u00faloh\u00e1ch, nap\u0159\u00edklad: zpracov\u00e1n\u00ed p\u0159eru\u0161en\u00ed, vstup\/v\u00fdstup, pr\u00e1ce se s\u00edt\u00ed atd. Hodnota m\u016f\u017ee b\u00fdt vysok\u00e1, pokud m\u00e1 VM velk\u00fd vstup\/v\u00fdstup.<\/p>\n<p><b>%OVRLP<\/b> \u2013 kolik \u010dasu fyzick\u00e9 j\u00e1dro, na kter\u00e9m b\u011b\u017e\u00ed VM proces, str\u00e1vilo na \u00faloh\u00e1ch jin\u00fdch proces\u016f.<\/p>\n<p>Tato m\u011b\u0159en\u00ed jsou mezi sebou propojena n\u00e1sleduj\u00edc\u00edm zp\u016fsobem:<\/p>\n<p>%USED = %RUN + %SYS \u2014 %OVRLP.<\/p>\n<p>Obvykle je metrika %USED informativn\u011bj\u0161\u00ed.<\/p>\n<p><b>%WAIT<\/b> \u2013 kolik \u010dasu byl proces ve stavu Wait b\u011bhem sledovan\u00e9ho obdob\u00ed. Zahrnuje IDLE.<\/p>\n<p><b>%IDLE<\/b> \u2013 kolik \u010dasu byl proces ve stavu IDLE b\u011bhem sledovan\u00e9ho obdob\u00ed.<\/p>\n<p><b>%SWPWT<\/b> \u2013 kolik \u010dasu v pr\u016fb\u011bhu sledovan\u00e9ho obdob\u00ed vCPU \u010dekal na operaci VMkernel Swap.<\/p>\n<p><b>%VMWAIT<\/b> \u2013 kolik \u010dasu byl vCPU ve stavu \u010dek\u00e1n\u00ed na ud\u00e1lost (obvykle vstup\/v\u00fdstup) b\u011bhem sledovan\u00e9ho obdob\u00ed. V vCenter nen\u00ed ekvivalentn\u00ed \u010d\u00edta\u010d. Vysok\u00e9 hodnoty nazna\u010duj\u00ed probl\u00e9my s vstupem\/v\u00fdstupem na VM.<\/p>\n<p>%WAIT = %VMWAIT + %IDLE + %SWPWT.<\/p>\n<p>Pokud VM nepou\u017e\u00edv\u00e1 VMkernel Swap, je p\u0159i anal\u00fdze probl\u00e9m\u016f s v\u00fdkonem rozumn\u00e9 se zam\u011b\u0159it na %VMWAIT, proto\u017ee tato metrika nezohled\u0148uje \u010das, kdy VM nic ned\u011blala (%IDLE).<\/p>\n<p><b>%RDY<\/b> \u2013 kolik \u010dasu byl proces ve stavu Ready b\u011bhem sledovan\u00e9ho obdob\u00ed.<\/p>\n<p><b>%CSTP<\/b> \u2013 kolik \u010dasu byl proces ve stavu costop b\u011bhem sledovan\u00e9ho obdob\u00ed.<\/p>\n<p><b>%MLMTD<\/b> \u2013 kolik \u010dasu byl vCPU ve stavu Ready kv\u016fli nastaven\u00e9mu limitu na zdroje b\u011bhem sledovan\u00e9ho obdob\u00ed.<\/p>\n<p>%WAIT + %RDY + %CSTP + %RUN = 100% \u2013 j\u00e1dro VM je v\u017edy v n\u011bjak\u00e9m z t\u011bchto \u010dty\u0159 stav\u016f.<\/p>\n<h3>CPU na hypervizoru<\/h3>\n<p>\nV vCenter jsou tak\u00e9 \u010d\u00edta\u010de v\u00fdkonnosti CPU pro hypervizor, ale nep\u0159edstavuj\u00ed nic zaj\u00edmav\u00e9ho \u2013 jsou to prost\u011b sou\u010dty \u010d\u00edta\u010d\u016f ze v\u0161ech VM na serveru.<br \/>\nNejl\u00e9pe je sledovat stav CPU na serveru na z\u00e1lo\u017ece Shrnut\u00ed:<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/15b6b9a6da136d892fbaf7f80fc206c1.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nPro server, stejn\u011b jako pro virtu\u00e1ln\u00ed stroj, existuje standardn\u00ed Alarm:<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/c3ae19ce3acd55cfff40d71437ee3a08.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nP\u0159i vysok\u00e9 z\u00e1t\u011b\u017ei CPU serveru za\u010d\u00ednaj\u00ed VM, kter\u00e9 na n\u011bm b\u011b\u017e\u00ed, m\u00edt probl\u00e9my s v\u00fdkonem.<\/p>\n<p>V ESXTOP jsou data o zat\u00ed\u017een\u00ed CPU serveru zobrazena v horn\u00ed \u010d\u00e1sti obrazovky. Krom\u011b standardn\u00edho zat\u00ed\u017een\u00ed CPU, kter\u00e9 je m\u00e1lo informativn\u00ed pro hypervizory, existuj\u00ed je\u0161t\u011b t\u0159i metriky:<\/p>\n<p><b>CORE UTIL(%)<\/b> \u2013 zat\u00ed\u017een\u00ed j\u00e1dra fyzick\u00e9ho serveru. Tento m\u011b\u0159i\u010d ukazuje, kolik \u010dasu b\u011bhem m\u011b\u0159\u00edc\u00edho obdob\u00ed j\u00e1dro vykon\u00e1valo pr\u00e1ci.<\/p>\n<p><b>PCPU UTIL(%)<\/b> \u2013 pokud je aktivov\u00e1n hyper-threading, na ka\u017ed\u00e9 fyzick\u00e9 j\u00e1dro p\u0159ipadnou dva vl\u00e1kna (PCPU). Tato metrika ukazuje, kolik \u010dasu ka\u017ed\u00e9 vl\u00e1kno vykon\u00e1valo pr\u00e1ci.<\/p>\n<p><b>PCPU USED(%)<\/b> \u2013 tot\u00e9\u017e jako PCPU UTIL(%), ale bere v \u00favahu frequency scaling (bu\u010f sn\u00ed\u017een\u00ed frekvence j\u00e1dra za \u00fa\u010delem \u00faspory energie, nebo zv\u00fd\u0161en\u00ed frekvence j\u00e1dra pomoc\u00ed technologie Turbo Boost) a hyper-threading.<\/p>\n<p>PCPU_USED% = PCPU_UTIL% * efektivn\u00ed frekvence j\u00e1dra \/ nomin\u00e1ln\u00ed frekvence j\u00e1dra.<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/dd8562ea91bee55bbf630358df26c217.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<i>Na tomto sn\u00edmku obrazovky je pro n\u011bkter\u00e1 j\u00e1dra hodnoty USED vy\u0161\u0161\u00ed ne\u017e 100%, proto\u017ee frekvence j\u00e1dra je vy\u0161\u0161\u00ed ne\u017e nomin\u00e1ln\u00ed kv\u016fli pr\u00e1ci Turbo Boost.<\/i><\/p>\n<p>P\u00e1r slov o tom, jak se zohled\u0148uje hyper-threading. Pokud procesy b\u011b\u017e\u00ed 100% \u010dasu na obou vl\u00e1knech fyzick\u00e9ho j\u00e1dra serveru, p\u0159i\u010dem\u017e j\u00e1dro pracuje na nomin\u00e1ln\u00ed frekvenci, pak:<\/p>\n<ul>\n<li>CORE UTIL pro j\u00e1dro bude 100%,<\/li>\n<li>PCPU UTIL pro ob\u011b vl\u00e1kna bude 100%,<\/li>\n<li>PCPU USED pro ob\u011b vl\u00e1kna bude 50%.<\/li>\n<\/ul>\n<p>\nPokud ob\u011b vl\u00e1kna nepracovala 100% \u010dasu b\u011bhem m\u011b\u0159en\u00e9ho obdob\u00ed, pak se v t\u011bch obdob\u00edch, kdy vl\u00e1kna pracovaly sou\u010dasn\u011b, PCPU USED pro j\u00e1dra d\u011bl\u00ed na polovinu.<\/p>\n<p>V ESXTOP je tak\u00e9 obrazovka s parametry spot\u0159eby energie CPU serveru. Zde m\u016f\u017eete zjistit, zda server vyu\u017e\u00edv\u00e1 technologie \u00faspory energie: C-states a P-states. Otev\u0159e se kl\u00e1vesou \u201ep\u201c:<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/531bf847b99f86bec0bd90e005919c79.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<\/p>\n<h3>Standardn\u00ed probl\u00e9my s v\u00fdkonem CPU<\/h3>\n<p>\nNakonec si projdu typick\u00e9 p\u0159\u00ed\u010diny probl\u00e9m\u016f s v\u00fdkonem CPU VM a d\u00e1m kr\u00e1tk\u00e9 rady, jak je \u0159e\u0161it:<\/p>\n<p><b>Nedostatek taktovac\u00ed frekvence j\u00e1dra.<\/b> Pokud nen\u00ed mo\u017en\u00e9 p\u0159ev\u00e9st VM na v\u00fdkonn\u011bj\u0161\u00ed j\u00e1dra, m\u016f\u017eete zkusit zm\u011bnit nastaven\u00ed spr\u00e1vy energie, aby Turbo Boost fungoval efektivn\u011bji.<\/p>\n<p><b>Nespr\u00e1vn\u00e9 velikov\u00e1n\u00ed VM (p\u0159\u00edli\u0161 mnoho\/m\u00e1lo jader).<\/b> Pokud je jader m\u00e1lo, bude vysok\u00e9 zat\u00ed\u017een\u00ed CPU VM. Pokud hodn\u011b, naraz\u00edte na vysok\u00fd co-stop.<\/p>\n<p><b>Velk\u00e1 p\u0159eplatba za CPU na serveru.<\/b> Pokud m\u00e1 VM vysok\u00fd Ready, sni\u017ete p\u0159eplatbu za CPU.<\/p>\n<p><b>Nespr\u00e1vn\u00e1 NUMA topologie na velk\u00fdch VM.<\/b> Topologie NUMA, kterou vid\u00ed VM (vNUMA), by m\u011bla odpov\u00eddat topologii NUMA serveru (pNUMA). O diagnostice a mo\u017en\u00fdch \u0159e\u0161en\u00edch tohoto probl\u00e9mu se p\u00ed\u0161e nap\u0159\u00edklad v knize <noindex><a rel=\"nofollow\" href=\"https:\/\/pages.rubrik.com\/host-resources-deep-dive_request.html\">\u201eVMware vSphere 6.5 Host Resources Deep Dive\u201c<\/a><\/noindex>. Pokud se nechcete p\u0159\u00edli\u0161 zab\u00fdvat a nem\u00e1te licen\u010dn\u00ed omezen\u00ed na OS nainstalovan\u00e9m na VM, vytvo\u0159te na VM v\u00edce virtu\u00e1ln\u00edch soket\u016f s jedn\u00edm j\u00e1drem. Mnoho neztrat\u00edte \ud83d\ude42<\/p>\n<p>To je v\u0161e ohledn\u011b CPU. Ptejte se. V dal\u0161\u00ed \u010d\u00e1sti v\u00e1m pov\u00edm o opera\u010dn\u00ed pam\u011bti.<\/p>\n<p><b class=\"spoiler_title\">U\u017eite\u010dn\u00e9 odkazy<\/b><noindex><a rel=\"nofollow\" href=\"http:\/\/virtual-red-dot.info\/vm-cpu-counters-vsphere\/\">http:\/\/virtual-red-dot.info\/vm-cpu-counters-vsphere\/<\/a><\/noindex><br \/>\n<noindex><a rel=\"nofollow\" href=\"https:\/\/kb.vmware.com\/kb\/1017926\">https:\/\/kb.vmware.com\/kb\/1017926<\/a><\/noindex><br \/>\n<noindex><a rel=\"nofollow\" href=\"http:\/\/www.yellow-bricks.com\/2012\/07\/17\/why-is-wait-so-high\/\">http:\/\/www.yellow-bricks.com\/2012\/07\/17\/why-is-wait-so-high\/<\/a><\/noindex><br \/>\n<noindex><a rel=\"nofollow\" href=\"https:\/\/communities.vmware.com\/docs\/DOC-9279\">https:\/\/communities.vmware.com\/docs\/DOC-9279<\/a><\/noindex><br \/>\n<noindex><a rel=\"nofollow\" href=\"https:\/\/www.vmware.com\/content\/dam\/digitalmarketing\/vmware\/en\/pdf\/techpaper\/performance\/whats-new-vsphere65-perf.pdf\">https:\/\/www.vmware.com\/content\/dam\/digitalmarketing\/vmware\/en\/pdf\/techpaper\/performance\/whats-new-vsphere65-perf.pdf<\/a><\/noindex><br \/>\n<noindex><a rel=\"nofollow\" href=\"https:\/\/pages.rubrik.com\/host-resources-deep-dive_request.html\">https:\/\/pages.rubrik.com\/host-resources-deep-dive_request.html<\/a><\/noindex><br \/>\n<br \/>Zdroj: <a content=\"nofollow\" rel=\"nofollow\" href=\"https:\/\/habr.com\/ru\/company\/dataline\/blog\/452884\/\">habr.com<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"<p>\u0415\u0441\u043b\u0438 \u0432\u044b \u0430\u0434\u043c\u0438\u043d\u0438\u0441\u0442\u0440\u0438\u0440\u0443\u0435\u0442\u0435 \u0432\u0438\u0440\u0442\u0443\u0430\u043b\u044c\u043d\u0443\u044e \u0438\u043d\u0444\u0440\u0430\u0441\u0442\u0440\u0443\u043a\u0442\u0443\u0440\u0443 \u043d\u0430 \u0431\u0430\u0437\u0435 VMware vSphere (\u0438\u043b\u0438 \u043b\u044e\u0431\u043e\u0433\u043e \u0434\u0440\u0443\u0433\u043e\u0433\u043e \u0441\u0442\u0435\u043a\u0430 \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0439), \u0442\u043e \u043d\u0430\u0432\u0435\u0440\u043d\u044f\u043a\u0430 \u0447\u0430\u0441\u0442\u043e \u0441\u043b\u044b\u0448\u0438\u0442\u0435 \u043e\u0442 \u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u0442\u0435\u043b\u0435\u0439 \u0436\u0430\u043b\u043e\u0431\u044b: \u00ab\u0412\u0438\u0440\u0442\u0443\u0430\u043b\u044c\u043d\u0430\u044f \u043c\u0430\u0448\u0438\u043d\u0430 \u0440\u0430\u0431\u043e\u0442\u0430\u0435\u0442 \u043c\u0435\u0434\u043b\u0435\u043d\u043d\u043e!\u00bb. \u0412 \u044d\u0442\u043e\u043c \u0446\u0438\u043a\u043b\u0435 \u0441\u0442\u0430\u0442\u0435\u0439 \u0440\u0430\u0437\u0431\u0435\u0440\u0443 \u043c\u0435\u0442\u0440\u0438\u043a\u0438 \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0441\u0442\u0438 \u0438 \u0440\u0430\u0441\u0441\u043a\u0430\u0436\u0443, \u0447\u0442\u043e \u0438 \u043f\u043e\u0447\u0435\u043c\u0443 \u00ab\u0442\u043e\u0440\u043c\u043e\u0437\u0438\u0442\u00bb \u0438 \u043a\u0430\u043a \u0441\u0434\u0435\u043b\u0430\u0442\u044c \u0442\u0430\u043a, \u0447\u0442\u043e\u0431\u044b \u043d\u0435 \u00ab\u0442\u043e\u0440\u043c\u043e\u0437\u0438\u043b\u043e\u00bb. \u0411\u0443\u0434\u0443 \u0440\u0430\u0441\u0441\u043c\u0430\u0442\u0440\u0438\u0432\u0430\u0442\u044c \u0441\u043b\u0435\u0434\u0443\u044e\u0449\u0438\u0435 \u0430\u0441\u043f\u0435\u043a\u0442\u044b \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0441\u0442\u0438 \u0432\u0438\u0440\u0442\u0443\u0430\u043b\u044c\u043d\u044b\u0445 \u043c\u0430\u0448\u0438\u043d: CPU, RAM, DISK, [&hellip;]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":25892,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[688],"tags":[],"class_list":["post-34335","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-administrirovanie"],"aioseo_notices":[],"aioseo_head":"\n\t\t<!-- All in One SEO 5.0.3 - aioseo.com -->\n\t<meta name=\"robots\" content=\"max-image-preview:large\" \/>\n\t<meta name=\"author\" content=\"Yuri Gagarin\"\/>\n\t<link rel=\"canonical\" href=\"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/analiz-proizvoditelnosti-virtualnoj-mashiny-v-vmware-vsphere-chast-1-cpu\" \/>\n\t\t<meta name=\"generator\" content=\"All in One SEO (AIOSEO) 5.0.3\" \/>\n\t\t<meta property=\"og:locale\" content=\"cs_CZ\" \/>\n\t\t<meta property=\"og:site_name\" content=\"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\" \/>\n\t\t<meta property=\"og:type\" content=\"article\" \/>\n\t\t<meta property=\"og:title\" content=\"\ud83e\udd47\u0410\u043d\u0430\u043b\u0438\u0437 \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0441\u0442\u0438 \u0432\u0438\u0440\u0442\u0443\u0430\u043b\u044c\u043d\u043e\u0439 \u043c\u0430\u0448\u0438\u043d\u044b \u0432 VMware vSphere. \u0427\u0430\u0441\u0442\u044c 1: CPU | ProHoster\" \/>\n\t\t<meta property=\"og:url\" content=\"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/analiz-proizvoditelnosti-virtualnoj-mashiny-v-vmware-vsphere-chast-1-cpu\" \/>\n\t\t<meta property=\"og:image\" content=\"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg\" \/>\n\t\t<meta property=\"og:image:secure_url\" content=\"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg\" \/>\n\t\t<meta property=\"og:image:width\" content=\"350\" \/>\n\t\t<meta property=\"og:image:height\" content=\"350\" \/>\n\t\t<meta property=\"article:published_time\" content=\"2019-10-31T18:57:43+00:00\" \/>\n\t\t<meta property=\"article:modified_time\" content=\"2019-10-31T18:57:43+00:00\" \/>\n\t\t<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/prohoster\" \/>\n\t\t<meta property=\"article:author\" content=\"https:\/\/www.facebook.com\/prohoster\" \/>\n\t\t<!-- All in One SEO -->\n\n","aioseo_head_json":{"title":"\ud83e\udd47Anal\u00fdza v\u00fdkonu virtu\u00e1ln\u00edho stroje ve VMware vSphere. \u010c\u00e1st 1: CPU | ProHoster","description":"","canonical_url":"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/analiz-proizvoditelnosti-virtualnoj-mashiny-v-vmware-vsphere-chast-1-cpu","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\u0410\u043d\u0430\u043b\u0438\u0437 \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0441\u0442\u0438 \u0432\u0438\u0440\u0442\u0443\u0430\u043b\u044c\u043d\u043e\u0439 \u043c\u0430\u0448\u0438\u043d\u044b \u0432 VMware vSphere. \u0427\u0430\u0441\u0442\u044c 1: CPU | ProHoster","og:url":"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/analiz-proizvoditelnosti-virtualnoj-mashiny-v-vmware-vsphere-chast-1-cpu","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":"2019-10-31T18:57:43+00:00","article:modified_time":"2019-10-31T18:57:43+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"34335","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":"2026-01-21 18:48:19","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2021-03-01 02:23:17","updated":"2026-01-21 18:48:19","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\/34335","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=34335"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts\/34335\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/media\/25892"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/media?parent=34335"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/categories?post=34335"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/tags?post=34335"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}