{"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\/sk\/blog\/administrirovanie\/analiz-proizvoditelnosti-virtualnoj-mashiny-v-vmware-vsphere-chast-1-cpu","title":{"rendered":"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/0c077a1754c2e32c1f339999ceacf03a.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nAk spravujete virtu\u00e1lnu infra\u0161trukt\u00faru zalo\u017een\u00fa na VMware vSphere (alebo akomko\u013evek inom technologickom stacku), ur\u010dite \u010dasto po\u010dujete od pou\u017e\u00edvate\u013eov s\u0165a\u017enosti: \u201eVirtu\u00e1lny stroj funguje pomaly!\u201c. V tejto s\u00e9rii \u010dl\u00e1nkov sa zameriam na metriky v\u00fdkonu a vysvetl\u00edm, \u010do a pre\u010do \u201espoma\u013euje\u201c a ako to spr\u00e1vne nastavi\u0165, aby to \u201enespomalilo\u201c.<\/p>\n<p>Bud\u00fa sa analyzova\u0165 nasleduj\u00face aspekty v\u00fdkonu virtu\u00e1lnych strojov:<\/p>\n<ul>\n<li>CPU,<\/li>\n<li>RAM,<\/li>\n<li>DISK,<\/li>\n<li>Sie\u0165.<\/li>\n<\/ul>\n<p>\nZa\u010dnem s CPU.<\/p>\n<p>Na anal\u00fdzu v\u00fdkonu budeme potrebova\u0165:<\/p>\n<ul>\n<li><b>vCenter Performance Counters<\/b> \u2013 v\u00fdkonnostn\u00e9 \u0161tatistiky, ktor\u00fdch grafy si m\u00f4\u017eete pozrie\u0165 prostredn\u00edctvom vSphere Client. Inform\u00e1cie o t\u00fdchto \u0161tatistik\u00e1ch s\u00fa dostupn\u00e9 v akejko\u013evek verzii klienta (\"hrub\u00fd\" klient na C#, webov\u00fd klient na Flex a webov\u00fd klient na HTML5). V t\u00fdchto \u010dl\u00e1nkoch budeme pou\u017e\u00edva\u0165 sn\u00edmky obrazovky z C# klienta, len preto, \u017ee vyzeraj\u00fa lep\u0161ie v miniat\u00fare:)<\/li>\n<li><b>ESXTOP<\/b> \u2013 n\u00e1stroj, ktor\u00fd sa sp\u00fa\u0161\u0165a z pr\u00edkazov\u00e9ho riadku ESXi. S jeho pomocou m\u00f4\u017eete z\u00edska\u0165 hodnoty v\u00fdkonov\u00fdch \u0161tatist\u00edk v re\u00e1lnom \u010dase alebo exportova\u0165 tieto hodnoty za ur\u010dit\u00fd \u010das do .csv s\u00faboru na \u010fal\u0161iu anal\u00fdzu. N\u00e1sledne poskytnem podrobnej\u0161ie inform\u00e1cie o tomto n\u00e1stroji a prinesiem nieko\u013eko u\u017eito\u010dn\u00fdch odkazov na dokument\u00e1ciu a \u010dl\u00e1nky na t\u00e9mu.<\/li>\n<\/ul>\n<p>\n<noindex><a rel=\"nofollow\" name=\"habracut\"><\/a><\/noindex><\/p>\n<h3>Trochu te\u00f3rie<\/h3>\n<p>\n<img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/bc415aace7653c45850d26b7104483c9.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nV ESXi m\u00e1 ka\u017ed\u00fd vCPU (jadro virtu\u00e1lneho stroja) pridelen\u00fd samostatn\u00fd proces \u2013 svet v terminol\u00f3gii VMware. Existuj\u00fa tie\u017e pomocn\u00e9 procesy, ale z h\u013eadiska anal\u00fdzy v\u00fdkonu VM s\u00fa menej zauj\u00edmav\u00e9.<\/p>\n<p>Proces v ESXi m\u00f4\u017ee by\u0165 v jednom z \u0161tyroch stavov:<\/p>\n<ul>\n<li><b>Spusti\u0165<\/b> \u2013 proces vykon\u00e1va nejak\u00fa u\u017eito\u010dn\u00fa pr\u00e1cu.<\/li>\n<li><b>\u010caka\u0165<\/b> \u2013 proces nevykon\u00e1va \u017eiadnu pr\u00e1cu (ne\u010dinn\u00fd) alebo \u010dak\u00e1 na vstup\/v\u00fdstup.<\/li>\n<li><b>N\u00e1klady<\/b> \u2013 stav, ktor\u00fd vznik\u00e1 v viacjadrov\u00fdch virtu\u00e1lnych strojoch. Vznik\u00e1, ke\u010f pl\u00e1nova\u010d CPU hyperv\u00edzora (ESXi CPU Scheduler) nedok\u00e1\u017ee napl\u00e1nova\u0165 s\u00fabe\u017en\u00e9 vykon\u00e1vanie v\u0161etk\u00fdch akt\u00edvnych jadier virtu\u00e1lneho stroja na fyzick\u00fdch jadr\u00e1ch servera. V re\u00e1lnom svete funguj\u00fa v\u0161etky jadr\u00e1 procesora paralelne, hos\u0165uj\u00faci opera\u010dn\u00fd syst\u00e9m vo VM po\u010d\u00edta s podobn\u00fdm spr\u00e1van\u00edm, preto mus\u00ed hyperv\u00edzor spomali\u0165 jadr\u00e1 VM, ktor\u00e9 maj\u00fa schopnos\u0165 ukon\u010di\u0165 takt r\u00fdchlej\u0161ie. V modern\u00fdch verzi\u00e1ch ESXi pou\u017e\u00edvate\u013e pl\u00e1nova\u010d CPU mechanizmus, ktor\u00fd sa naz\u00fdva relaxed co-scheduling: hyperv\u00edzor zoh\u013ead\u0148uje odch\u00fdlku medzi najr\u00fdchlej\u0161\u00edm a najpomal\u0161\u00edm jadrom virtu\u00e1lneho stroja (skew). Ak odch\u00fdlka presiahne ur\u010dit\u00fd prah, \u201er\u00fdchle\u201c jadro prech\u00e1dza do stavu costop. Ak jadr\u00e1 VM tr\u00e1via ve\u013ea \u010dasu v tomto stave, m\u00f4\u017ee to sp\u00f4sobi\u0165 probl\u00e9my s v\u00fdkonnos\u0165ou.<\/li>\n<li><b>Pripraven\u00e9<\/b> \u2013 proces prech\u00e1dza do tohto stavu, ke\u010f hyperv\u00edzor nem\u00e1 mo\u017enos\u0165 priradi\u0165 zdroje na jeho vykonanie. Vysok\u00e9 hodnoty ready m\u00f4\u017eu sp\u00f4sobi\u0165 probl\u00e9my s v\u00fdkonom VM.<\/li>\n<\/ul>\n<p><\/p>\n<h3>Hlavn\u00e9 ukazovatele v\u00fdkonu CPU virtu\u00e1lneho stroja<\/h3>\n<p>\n<b>Vyu\u017eitie CPU, %.<\/b> Zobrazuje percento vyu\u017eitia CPU za stanoven\u00e9 obdobie.<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/7a5da7dcd45329c3aae8e394c5d35a30.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\n<b>Ako analyzova\u0165?<\/b> Ak VM stabilne vyu\u017e\u00edva CPU na 90 % alebo s\u00fa v\u00fdkyvy a\u017e do 100 %, m\u00e1me probl\u00e9m. Probl\u00e9my sa nemusia prejavova\u0165 len v \"pomaly\" fungovan\u00ed aplik\u00e1cie vo VM, ale aj v nedostupnosti VM cez sie\u0165. Ak monitorovac\u00ed syst\u00e9m ukazuje, \u017ee VM ob\u010das vypad\u00e1va, venujte pozornos\u0165 v\u00fdkyvom na grafe CPU Usage.<\/p>\n<p>Existuje \u0161tandardn\u00fd Alarm, ktor\u00fd zobrazuje za\u0165a\u017eenie CPU virtu\u00e1lneho stroja:<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/9bfd9bb5b43b9b02e8bbfac307d6c9f9.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\n<b>\u010co robi\u0165?<\/b> Ak CPU Usage vo VM neust\u00e1le presahuje normu, m\u00f4\u017eete zv\u00e1\u017ei\u0165 zv\u00fd\u0161enie po\u010dtu vCPU (bohu\u017eia\u013e, to nie v\u017edy pom\u00f4\u017ee) alebo presun\u00fa\u0165 VM na server s v\u00fdkonnej\u0161\u00edmi procesormi.<\/p>\n<h3>Vyu\u017eitie CPU v MHz<\/h3>\n<p>\nNa grafoch v vCenter Usage v % m\u00f4\u017eete pozera\u0165 len celkov\u00fd pre cel\u00fd virtu\u00e1lny stroj, grafy pod\u013ea jednotliv\u00fdch jadier nie s\u00fa dostupn\u00e9 (v Esxtop s\u00fa hodnoty v % pod\u013ea jadier). Pre ka\u017ed\u00e9 jadro si m\u00f4\u017eete pozrie\u0165 Vyu\u017eitie v MHz.<\/p>\n<p><b>Ako analyzova\u0165?<\/b> Niekedy sa st\u00e1va, \u017ee aplik\u00e1cia nie je optimalizovan\u00e1 pre viacjadrov\u00fa architekt\u00faru: vyu\u017e\u00edva na 100 % iba jedno jadro, zatia\u013e \u010do ostatn\u00e9 zost\u00e1vaj\u00fa nevy\u0165a\u017een\u00e9. Napr\u00edklad pri predvolen\u00fdch nastaveniach z\u00e1lohovania MS SQL sp\u00fa\u0161\u0165a proces iba na jednom jadre. V\u00fdsledkom je, \u017ee z\u00e1lohovanie spoma\u013euje nie kv\u00f4li pomalej r\u00fdchlosti diskov (na \u010do sa p\u00f4vodne s\u0165a\u017eoval pou\u017e\u00edvate\u013e), ale preto, \u017ee procesor to nezvl\u00e1da. Probl\u00e9m bol vyrie\u0161en\u00fd zmenou parametrov: z\u00e1lohovanie za\u010dalo prebieha\u0165 paralelne do nieko\u013ek\u00fdch s\u00faborov (pr\u00edslu\u0161ne do nieko\u013ek\u00fdch procesov).<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/d1bce5fe5ef89a69c31a6a8f043d5d24.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<i>Pr\u00edklad nerovnomern\u00e9ho za\u0165a\u017eenia jadier.<\/i><\/p>\n<p>Taktie\u017e sa m\u00f4\u017ee vyskytn\u00fa\u0165 situ\u00e1cia (ako na grafe vy\u0161\u0161ie), ke\u010f s\u00fa jadr\u00e1 za\u0165a\u017een\u00e9 nerovnomerne a na niektor\u00fdch z nich s\u00fa vrcholy na 100 %. Tak ako pri za\u0165a\u017een\u00ed iba jedn\u00e9ho jadra, alarm na CPU Usage sa neaktivuje (plat\u00ed pre cel\u00fa VM), ale probl\u00e9my s v\u00fdkonom bud\u00fa.<\/p>\n<p><b>\u010co robi\u0165? <\/b>Ak softv\u00e9r vo virtu\u00e1lnej ma\u0161ine za\u0165a\u017euje jadr\u00e1 nerovnomerne (vyu\u017e\u00edva len jedno jadro alebo \u010das\u0165 jadier), nem\u00e1 zmysel zvy\u0161ova\u0165 ich po\u010det. V takom pr\u00edpade je lep\u0161ie presun\u00fa\u0165 VM na server s v\u00fdkonnej\u0161\u00edmi procesormi.<\/p>\n<p>Taktie\u017e m\u00f4\u017eete sk\u00fasi\u0165 skontrolova\u0165 nastavenia energetickej \u00fa\u010dinnosti v BIOS servera. Mnoh\u00ed administr\u00e1tori aktivuj\u00fa v BIOSe re\u017eim Vysok\u00fd v\u00fdkon a t\u00fdmto sp\u00f4sobom vyp\u00ednaj\u00fa technol\u00f3gie energetickej \u00faspory C-states a P-states. V modern\u00fdch procesoroch Intel sa pou\u017e\u00edva technol\u00f3gia Turbo Boost, ktor\u00e1 zvy\u0161uje frekvenciu jednotliv\u00fdch jadier procesora na \u00fakor ostatn\u00fdch jadier. Ale funguje to len pri aktivovan\u00fdch technol\u00f3gi\u00e1ch energetickej \u00faspory. Ak ich vypneme, procesor nedok\u00e1\u017ee zn\u00ed\u017ei\u0165 energetick\u00fa spotrebu jadier, ktor\u00e9 nie s\u00fa za\u0165a\u017een\u00e9. <\/p>\n<p>VMware odpor\u00fa\u010da nevyp\u00edna\u0165 technol\u00f3gie energetickej \u00faspory na serveroch, ale vybra\u0165 re\u017eimy, ktor\u00e9 maxim\u00e1lne odovzdaj\u00fa kontrolu nad energetickou spotrebou hyperv\u00edzoru. Pri tomto by ste mali v nastaveniach energetickej \u00fa\u010dinnosti hyperv\u00edzora zvoli\u0165 Vysok\u00fd v\u00fdkon. <\/p>\n<p>Ak vo va\u0161ej infra\u0161trukt\u00fare jednotliv\u00e9 VM (alebo jadr\u00e1 VM) vy\u017eaduj\u00fa zv\u00fd\u0161en\u00fa frekvenciu CPU, spr\u00e1vne nastavenie energetickej spotreby m\u00f4\u017ee v\u00fdznamne zlep\u0161i\u0165 ich v\u00fdkon.<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/59aec2b77fa87e413d90371a11388629.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<\/p>\n<h3>CPU Ready (Pripravenos\u0165) <\/h3>\n<p>\nAk je jadro VM (vCPU) v stave Ready, nevykon\u00e1va \u017eiadnu u\u017eito\u010dn\u00fa pr\u00e1cu. Tento stav nast\u00e1va, ke\u010f hyperv\u00edzor nen\u00e1jde vo\u013en\u00e9 fyzick\u00e9 jadro, na ktor\u00e9 by mohol priradi\u0165 proces vCPU virtu\u00e1lnej stroje.<\/p>\n<p><b>Ako analyzova\u0165?<\/b> Ak s\u00fa jadra virtu\u00e1lnej stroje v stave Ready viac ako 10% \u010dasu, pravdepodobne si v\u0161imnete probl\u00e9my s v\u00fdkonom. Jednoducho povedan\u00e9, viac ako 10% \u010dasu VM \u010dak\u00e1 na dostupnos\u0165 fyzick\u00fdch zdrojov.<\/p>\n<p>V vCenter m\u00f4\u017eete vidie\u0165 2 po\u010d\u00edtadl\u00e1 s\u00favisiace s CPU Ready:<\/p>\n<ul>\n<li>Readiness,<\/li>\n<li>Ready.<\/li>\n<\/ul>\n<p>\nHodnoty oboch po\u010d\u00edtadiel je mo\u017en\u00e9 zobrazi\u0165 ako pre cel\u00fa VM, tak aj pre jednotliv\u00e9 jadr\u00e1.<br \/>\nReadiness ukazuje hodnotu ihne\u010f v percent\u00e1ch, ale iba v re\u00e1lnom \u010dase (d\u00e1ta za posledn\u00fa hodinu, interval merania 20 sek\u00fand). Toto po\u010d\u00edtadlo je najlep\u0161ie pou\u017e\u00edva\u0165 len pre okam\u017eit\u00e9 h\u013eadanie probl\u00e9mov.<\/p>\n<p>Hodnoty po\u010d\u00edtadla Ready je tie\u017e mo\u017en\u00e9 vidie\u0165 v historickej perspekt\u00edve. To je u\u017eito\u010dn\u00e9 pre zistenie vzorov a pre hlb\u0161iu anal\u00fdzu probl\u00e9mu. Napr\u00edklad, ak virtu\u00e1lna stroj m\u00e1 probl\u00e9my s v\u00fdkonom v ur\u010ditom \u010dase, m\u00f4\u017eeme porovna\u0165 intervaly zv\u00fd\u0161en\u00fdch hodn\u00f4t CPU Ready s celkovou z\u00e1\u0165a\u017eou na serveri, na ktorom dan\u00e1 VM be\u017e\u00ed, a prija\u0165 opatrenia na zn\u00ed\u017eenie z\u00e1\u0165a\u017ee (ak DRS nezvl\u00e1dol).<\/p>\n<p>Ready na rozdiel od Readiness nie je zobrazovan\u00fd v percent\u00e1ch, ale v milisekund\u00e1ch. Toto po\u010d\u00edtadlo typu Summation zobrazuje, ko\u013eko \u010dasu po\u010das meracieho obdobia jadro VM str\u00e1vilo v stave Ready. Tento \u00fadaj m\u00f4\u017eeme previes\u0165 na percent\u00e1 pomocou jednoduchej formul\u00e1cie:<\/p>\n<p>(CPU ready summation value \/ (default update interval grafu v sekund\u00e1ch * 1000)) * 100 = CPU ready %<\/p>\n<p>Napr\u00edklad pre VM na grafe ni\u017e\u0161ie dosiahneme vrcholov\u00fa hodnotu Ready pre cel\u00fa virtu\u00e1lnu stroj nasledovne: <\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 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\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/f0371c9cd655c1e4785e90debde61dc8.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nPri v\u00fdpo\u010dte hodnoty Ready v percent\u00e1ch je potrebn\u00e9 venova\u0165 pozornos\u0165 dvom aspektom:<\/p>\n<ul>\n<li>Hodnota Ready pre cel\u00fa VM je s\u00fa\u010dtom Ready pre jadr\u00e1.<\/li>\n<li>Interval merania. Pre Re\u00e1lny \u010das \u2013 je to 20 sek\u00fand, a napr\u00edklad na dennej grafike \u2013 to je 300 sek\u00fand.<\/li>\n<\/ul>\n<p>\nPri akt\u00edvnom rie\u0161en\u00ed probl\u00e9mov je \u013eahk\u00e9 tieto jednoduch\u00e9 aspekty prehliadnu\u0165 a strati\u0165 cenn\u00fd \u010das na rie\u0161enie neexistuj\u00facich probl\u00e9mov. <\/p>\n<p>Vypo\u010d\u00edtame Ready na z\u00e1klade \u00fadajov z grafu ni\u017e\u0161ie. (324474\/(20*1000))*100 = 1622% pre cel\u00fd VM. Ak sa pozrieme na jadr\u00e1, vyzer\u00e1 to menej zle: 1622\/64 = 25% na jadro. V tomto pr\u00edpade je odhalenie n\u00e1strahy celkom jednoduch\u00e9: hodnota Ready je nerealistick\u00e1. Ale ak hovor\u00edme o 10\u201320% pre cel\u00fd VM s viacer\u00fdmi jadrami, m\u00f4\u017ee by\u0165 hodnota na ka\u017ed\u00e9 jadro v norm\u00e1lnych medziach.<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/703ab2b1f1fab8bda10aad781c2cef78.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\n<b>\u010co robi\u0165? <\/b>Vysok\u00e1 hodnota Ready nazna\u010duje, \u017ee serveru ch\u00fdbaj\u00fa procesorov\u00e9 zdroje na norm\u00e1lnu prev\u00e1dzku virtu\u00e1lnych strojov. V tejto situ\u00e1cii zost\u00e1va len zn\u00ed\u017ei\u0165 prebytok procesora (vCPU:pCPU). Je jasn\u00e9, \u017ee to mo\u017eno dosiahnu\u0165 zn\u00ed\u017een\u00edm parametrov existuj\u00facich VM alebo presunom niektor\u00fdch VM na in\u00e9 servery.<\/p>\n<h3>Co-stop<\/h3>\n<p>\n<b>Ako analyzova\u0165?<\/b> Tento \u010d\u00edta\u010d m\u00e1 taktie\u017e typ Summation a prepo\u010d\u00edtava sa na percent\u00e1 podobne ako Ready:<\/p>\n<p>(CPU co-stop summation value \/ (chart default update interval in seconds * 1000)) * 100 = CPU co-stop %<\/p>\n<p>Tu je tie\u017e potrebn\u00e9 venova\u0165 pozornos\u0165 po\u010dtu jadier na VM a intervalu merania.<br \/>\nV stave co-stop jadro nevykon\u00e1va u\u017eito\u010dn\u00fa pr\u00e1cu. Pri spr\u00e1vnom v\u00fdbere ve\u013ekosti VM a norm\u00e1lnom za\u0165a\u017een\u00ed servera by mal by\u0165 \u010d\u00edta\u010d so-stop bl\u00edzko nuly.<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/e226afbf91155c2d5d3fdc4571e7383a.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<i>V tomto pr\u00edpade je za\u0165a\u017eenie zjavne abnorm\u00e1lne :)<\/i><\/p>\n<p><b>\u010co robi\u0165?<\/b> Ak na jednom hyperv\u00edzore pracuje nieko\u013eko VM s ve\u013ek\u00fdm po\u010dtom jadier a existuje prebytok CPU, potom m\u00f4\u017ee \u010d\u00edta\u010d co-stop nar\u00e1s\u0165, \u010do povedie k probl\u00e9mom s v\u00fdkonom t\u00fdchto VM. <\/p>\n<p>Co-stop sa taktie\u017e zv\u00fd\u0161i, ak sa pre akt\u00edvne jadr\u00e1 jednej VM pou\u017e\u00edvaj\u00fa vl\u00e1kna na jednom fyzickom jadre servera so zapnut\u00fdm hyper-threadingom. Tak\u00e1to situ\u00e1cia m\u00f4\u017ee nasta\u0165, ak m\u00e1 VM viac jadier, ne\u017e je fyzicky dostupn\u00fdch na serveri, kde be\u017e\u00ed, alebo ak je pre VM zapnut\u00e9 nastavenie 'preferHT'. O tomto nastaven\u00ed si m\u00f4\u017eete pre\u010d\u00edta\u0165 <noindex><a rel=\"nofollow\" href=\"https:\/\/blogs.vmware.com\/vsphere\/2014\/03\/perferht-use-2.html\">tu<\/a><\/noindex>. <\/p>\n<p>Aby ste predi\u0161li probl\u00e9mom s v\u00fdkonom VM z d\u00f4vodu vysok\u00e9ho co-stop, vyberajte ve\u013ekos\u0165 VM v s\u00falade s odpor\u00fa\u010daniami v\u00fdrobcu softv\u00e9ru, ktor\u00fd funguje na tejto VM, a s mo\u017enostami fyzick\u00e9ho servera, kde VM be\u017e\u00ed. <\/p>\n<p>Nerobte rezervn\u00e9 jadra, m\u00f4\u017ee to sp\u00f4sobi\u0165 probl\u00e9my s v\u00fdkonom nielen samotnej VM, ale aj jej susedov na serveri.<\/p>\n<h3>\u010eal\u0161ie u\u017eito\u010dn\u00e9 metriky CPU<\/h3>\n<p>\n<b>Spusti\u0165<\/b> \u2013 ko\u013eko \u010dasu (ms) po\u010das meranej doby bol vCPU v stave RUN, teda fakticky vykon\u00e1val u\u017eito\u010dn\u00fa pr\u00e1cu.<\/p>\n<p><b>Idle<\/b> \u2013 ko\u013eko \u010dasu (ms) bol vCPU po\u010das merania v stave ne\u010dinnosti. Vysok\u00e9 hodnoty Idle nie s\u00fa probl\u00e9m, len malo vCPU \"ni\u010d robi\u0165\".<\/p>\n<p><b>\u010caka\u0165<\/b> \u2013 ko\u013eko \u010dasu (ms) bol vCPU po\u010das merania v stave Wait. Ke\u010f\u017ee tento \u010d\u00edta\u010d obsahuje IDLE, vysok\u00e9 hodnoty Wait tie\u017e neznamenaj\u00fa probl\u00e9my. Ak je v\u0161ak vysok\u00e9 Wait a IDLE je n\u00edzke, znamen\u00e1 to, \u017ee VM \u010dakala na dokon\u010denie oper\u00e1ci\u00ed vstupu\/v\u00fdstupu, \u010do m\u00f4\u017ee nazna\u010dova\u0165 probl\u00e9my so v\u00fdkonom pevn\u00e9ho disku alebo in\u00fdch virtu\u00e1lnych zariaden\u00ed VM.<\/p>\n<p><b>Maxim\u00e1lne obmedzen\u00e9<\/b> \u2013 ko\u013eko \u010dasu (ms) bol vCPU po\u010das merania v stave Ready kv\u00f4li nastaven\u00e9mu limitu pre zdroje. Ak je v\u00fdkon nepochopite\u013ene n\u00edzky, je u\u017eito\u010dn\u00e9 skontrolova\u0165 hodnotu tohto \u010d\u00edta\u010da a limit CPU v nastaveniach VM. VM m\u00f4\u017eu ma\u0165 nastaven\u00e9 limity, o ktor\u00fdch neviete. Napr\u00edklad sa to stane, ke\u010f bola VM sklonovan\u00e1 z \u0161abl\u00f3ny, na ktorej bol nastaven\u00fd limit CPU.<\/p>\n<p><b>\u010cakanie na swap<\/b> \u2013 ko\u013eko \u010dasu vCPU \u010dakalo na oper\u00e1cie s VMkernel Swap po\u010das merania. Ak s\u00fa hodnoty tohto \u010d\u00edta\u010da vy\u0161\u0161ie ako nula, VM m\u00e1 ur\u010dite probl\u00e9my s v\u00fdkonom. Podrobnej\u0161ie o SWAP sa budeme bavi\u0165 v \u010dl\u00e1nku o \u010d\u00edta\u010doch pam\u00e4te.<\/p>\n<h3>ESXTOP<\/h3>\n<p>\nAk s\u00fa v\u00fdkonnostn\u00e9 \u010d\u00edta\u010de v vCenter dobr\u00e9 na anal\u00fdzu historick\u00fdch \u00fadajov, operat\u00edvna anal\u00fdza probl\u00e9mov je lep\u0161ie vykon\u00e1va\u0165 v ESXTOP. Tu s\u00fa v\u0161etky hodnoty zobrazen\u00e9 pripraven\u00e9 (nie je potrebn\u00e9 ni\u010d prev\u00e1dza\u0165) a minim\u00e1lny merac\u00ed interval je 2 sekundy.<br \/>\nObrazovka ESXTOP pre CPU sa vyvol\u00e1va stla\u010den\u00edm kl\u00e1vesy \u201ec\u201c a vyzer\u00e1 takto:<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/f4f3611bc383007bdbaa0948001923c4.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nPre pohodlie m\u00f4\u017eete zanecha\u0165 iba procesy virtu\u00e1lnych strojov stla\u010den\u00edm Shift-V.<br \/>\nAby ste si mohli pozrie\u0165 metriky pre jednotliv\u00e9 jadr\u00e1 VM, stla\u010dte \u201ee\u201c a zadajte GID zauj\u00edmav\u00e9ho VM (30919 na sn\u00edmke obrazovky ni\u017e\u0161ie):<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/5f2c0226cee6e4e43128537bd4da843f.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nKr\u00e1tko prejdeme cez st\u013apce, ktor\u00e9 s\u00fa predvolen\u00e9. \u010eal\u0161ie st\u013apce m\u00f4\u017eete prida\u0165 stla\u010den\u00edm \u201ef\u201c.<\/p>\n<p><b>NWLD (Po\u010det svetov)<\/b> \u2013 po\u010det procesov v skupine. Aby ste rozbalili skupinu a videli metriky pre ka\u017ed\u00fd proces (napr\u00edklad pre ka\u017ed\u00e9 jadro viacjadrovej VM), stla\u010dte \u201ee\u201c. Ak je v skupine viac ako jeden proces, hodnoty metr\u00edk pre skupinu s\u00fa rovn\u00e9 s\u00fa\u010dtu metr\u00edk pre jednotliv\u00e9 procesy.<\/p>\n<p><b>%POU\u017dIT\u00c9<\/b> \u2013 ko\u013eko cyklov CPU server vyu\u017e\u00edva proces alebo skupina procesov.<\/p>\n<p><b>%PREV\u00c1DZKA<\/b> \u2013 ko\u013eko \u010dasu bol proces po\u010das meran\u00e9ho obdobia v stave RUN, t.j. vykon\u00e1val u\u017eito\u010dn\u00fa pr\u00e1cu. Rozdiel od %USED spo\u010d\u00edva v tom, \u017ee nezoh\u013ead\u0148uje hyper-threading, frekven\u010dn\u00e9 \u0161k\u00e1lovanie 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, napr\u00edklad: spracovanie preru\u0161enia, vstup\/v\u00fdstup, pr\u00e1ca so sie\u0165ou at\u010f. Hodnota m\u00f4\u017ee by\u0165 vysok\u00e1, ak m\u00e1 VM vysok\u00fd vstup\/v\u00fdstup.<\/p>\n<p><b>%OVRLP<\/b> \u2013 ko\u013eko \u010dasu fyzick\u00e9 jadro, na ktorom sa proces VM vykon\u00e1va, str\u00e1vilo na \u00faloh\u00e1ch in\u00fdch procesov.<\/p>\n<p>Tieto metriky s\u00fa vz\u00e1jomne prepojen\u00e9 nasleduj\u00facim sp\u00f4sobom:<\/p>\n<p>%USED = %RUN + %SYS \u2013 %OVRLP.<\/p>\n<p>Zvy\u010dajne je metrika %USED informat\u00edvnej\u0161ia.<\/p>\n<p><b>%WAIT<\/b> \u2013 ko\u013eko \u010dasu bol proces po\u010das meran\u00e9ho obdobia v stave Wait. Zoh\u013ead\u0148uje IDLE.<\/p>\n<p><b>%IDLE<\/b> \u2013 ko\u013eko \u010dasu bol proces po\u010das meran\u00e9ho obdobia v stave IDLE.<\/p>\n<p><b>%SWPWT<\/b> \u2013 ko\u013eko \u010dasu bol proces v priebehu merania vCPU v \u010dakan\u00ed na oper\u00e1ciu so VMkernel Swap.<\/p>\n<p><b>%VMWAIT<\/b> \u2013 ko\u013eko \u010dasu bol vCPU po\u010das meran\u00e9ho obdobia v stave \u010dakania na udalos\u0165 (zvy\u010dajne vstup\/v\u00fdstup). Neexistuje ekvivalentn\u00fd \u010d\u00edta\u010d v vCenter. Vysok\u00e9 hodnoty nazna\u010duj\u00fa probl\u00e9my s vstupom\/v\u00fdstupom na VM.<\/p>\n<p>%WAIT = %VMWAIT + %IDLE + %SWPWT.<\/p>\n<p>Ak VM nepou\u017e\u00edva VMkernel Swap, je pri anal\u00fdze probl\u00e9mov s v\u00fdkonnos\u0165ou rozumn\u00e9 pozera\u0165 sa na %VMWAIT, preto\u017ee t\u00e1to metrika nezoh\u013ead\u0148uje \u010das, ke\u010f VM ni\u010d nerobila (%IDLE).<\/p>\n<p><b>%RDY<\/b> \u2013 ko\u013eko \u010dasu bol proces po\u010das meran\u00e9ho obdobia v stave Ready.<\/p>\n<p><b>%CSTP<\/b> \u2013 ko\u013eko \u010dasu bol proces po\u010das meran\u00e9ho obdobia v stave costop.<\/p>\n<p><b>%MLMTD<\/b> \u2013 ko\u013eko \u010dasu bol vCPU po\u010das meran\u00e9ho obdobia v stave Ready z d\u00f4vodu nastaven\u00e9ho limitu na zdroje.<\/p>\n<p>%WAIT + %RDY + %CSTP + %RUN = 100% \u2013 jadro VM je st\u00e1le v jednom z t\u00fdchto \u0161tyroch stavov.<\/p>\n<h3>CPU na hyperv\u00edzore<\/h3>\n<p>\nV vCenter s\u00fa tie\u017e dostupn\u00e9 \u00farovne v\u00fdkonu CPU pre hyperv\u00edzor, ale nezoh\u013ead\u0148uj\u00fa ni\u010d zauj\u00edmav\u00e9 \u2013 je to len s\u00fa\u010det \u010d\u00edta\u010dov zo v\u0161etk\u00fdch VM na serveri.<br \/>\nNajpr\u00edjemnej\u0161ie je pozera\u0165 stav CPU na serveri na z\u00e1lo\u017eke Preh\u013ead:<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/15b6b9a6da136d892fbaf7f80fc206c1.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nPre server, rovnako ako pre virtu\u00e1lny stroj, existuje \u0161tandardn\u00e1 v\u00fdstraha:<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/c3ae19ce3acd55cfff40d71437ee3a08.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nPri vysokom za\u0165a\u017een\u00ed CPU servera sa VM, ktor\u00e9 na \u0148om pracuj\u00fa, stret\u00e1vaj\u00fa s probl\u00e9mami s v\u00fdkonnos\u0165ou.<\/p>\n<p>V ESXTOP s\u00fa \u00fadaje o za\u0165a\u017een\u00ed CPU servera zobrazen\u00e9 v hornej \u010dasti obrazovky. Okrem \u0161tandardn\u00e9ho za\u0165a\u017eenia CPU, ktor\u00e9 je pre hyperv\u00edzory margin\u00e1lne informativne, s\u00fa tu e\u0161te tri metriky:<\/p>\n<p><b>CORE UTIL(%)<\/b> \u2013 za\u0165a\u017eenie jadra fyzick\u00e9ho servera. T\u00e1to metrika ukazuje, ko\u013eko \u010dasu po\u010das meracieho obdobia jadro vykon\u00e1valo pr\u00e1cu.<\/p>\n<p><b>PCPU UTIL(%)<\/b> \u2013 ak je aktivovan\u00fd hyper-threading, na ka\u017ed\u00e9 fyzick\u00e9 jadro pripadn\u00fa dva vl\u00e1kna (PCPU). T\u00e1to metrika ukazuje, ko\u013eko \u010dasu ka\u017ed\u00e9 vl\u00e1kno vykon\u00e1valo pr\u00e1cu.<\/p>\n<p><b>PCPU USED(%)<\/b> \u2013 to ist\u00e9 ako PCPU UTIL(%), ale berie do \u00favahy frekven\u010dn\u00e9 \u0161k\u00e1lovanie (bu\u010f zn\u00ed\u017eenie frekvencie jadra za \u00fa\u010delom \u00faspory energie, alebo zv\u00fd\u0161enie frekvencie jadra technol\u00f3giou Turbo Boost) a hyper-threading.<\/p>\n<p>PCPU_USED% = PCPU_UTIL% * efekt\u00edvna frekvencia jadra \/ nomin\u00e1lna frekvencia jadra.<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/dd8562ea91bee55bbf630358df26c217.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<i>Na tomto sn\u00edmku obrazovky pre niektor\u00e9 jadr\u00e1 je z d\u00f4vodu fungovania Turbo Boost hodnota USED vy\u0161\u0161ia ako 100%, preto\u017ee frekvencia jadra je nad nomin\u00e1lnou.<\/i><\/p>\n<p>P\u00e1r slov o tom, ako sa zoh\u013ead\u0148uje hyper-threading. Ak procesy be\u017eia 100% \u010dasu na oboch vl\u00e1knach fyzick\u00e9ho jadra servera a jadro pracuje na nomin\u00e1lnej frekvencii, tak:<\/p>\n<ul>\n<li>CORE UTIL pre jadro bude 100%,<\/li>\n<li>PCPU UTIL pre obe vl\u00e1kna bude 100%,<\/li>\n<li>PCPU USED pre obe vl\u00e1kna bude 50%.<\/li>\n<\/ul>\n<p>\nAk oba vl\u00e1kna nefungovali 100% \u010dasu po\u010das meracieho obdobia, v t\u00fdch obdobiach, ke\u010f vl\u00e1kna pracovali paralelne, sa PCPU USED pre jadr\u00e1 del\u00ed na polovicu.<\/p>\n<p>V ESXTOP je tie\u017e obrazovka s parametrami spotreby energie CPU servera. Tu si m\u00f4\u017eete skontrolova\u0165, \u010di server pou\u017e\u00edva technol\u00f3gie na \u00fasporu energie: C-states a P-states. Aktivuje sa kl\u00e1vesom \u201ep\u201c:<\/p>\n<p><img decoding=\"async\" alt=\"Anal\u00fdza v\u00fdkonu virtu\u00e1lneho stroja vo VMware vSphere. \u010cas\u0165 1: CPU\" src=\"\/wp-content\/uploads\/2019\/05\/531bf847b99f86bec0bd90e005919c79.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<\/p>\n<h3>\u0160tandardn\u00e9 probl\u00e9my s v\u00fdkonom CPU<\/h3>\n<p>\nNa z\u00e1ver sa pozriem na typick\u00e9 pr\u00ed\u010diny probl\u00e9mov s v\u00fdkonom CPU VM a poskytnem kr\u00e1tke rady na ich rie\u0161enie:<\/p>\n<p><b>Nedostatok taktn\u00fdch frekvenci\u00ed jadra.<\/b> Ak nie je mo\u017en\u00e9 prenies\u0165 VM na v\u00fdkonnej\u0161ie jadr\u00e1, m\u00f4\u017eete sa pok\u00fasi\u0165 zmeni\u0165 nastavenia energetickej spotreby, aby Turbo Boost fungoval efekt\u00edvnej\u0161ie.<\/p>\n<p><b>Nespr\u00e1vne sizing VM (pr\u00edli\u0161 ve\u013ea\/m\u00e1lo jadier).<\/b> Ak nastav\u00edte m\u00e1lo jadier, nastane vysok\u00e9 za\u0165a\u017eenie CPU VM. Ak ich je ve\u013ea, zachyt\u00edte vysok\u00e9 co-stop.<\/p>\n<p><b>Ve\u013ek\u00e1 preplatba CPU na serveri.<\/b> Ak m\u00e1 VM vysok\u00fd Ready, zn\u00ed\u017ete preplatbu CPU.<\/p>\n<p><b>Nespr\u00e1vna NUMA topol\u00f3gia na ve\u013ek\u00fdch VM.<\/b> TOPOL\u00d3GIA NUMA, ktor\u00fa vid\u00ed VM (vNUMA), mus\u00ed zodpoveda\u0165 TOPOL\u00d3GII NUMA servera (pNUMA). O diagnostike a mo\u017en\u00fdch rie\u0161eniach tohto probl\u00e9mu sa p\u00ed\u0161e napr\u00edklad v knihe <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>. Ak sa nechcete hlboko zaobera\u0165 a nem\u00e1te licen\u010dn\u00e9 obmedzenia pre OS nain\u0161talovan\u00fd na VM, vytvorte na VM ve\u013ea virtu\u00e1lnych soketov s jednou jadrom. Mnoho ste nestratili \ud83d\ude42<\/p>\n<p>Toto je v\u0161etko o CPU. Kl\u00e1s\u0165 ot\u00e1zky. V nasleduj\u00facej \u010dasti poviem o opera\u010dnej pam\u00e4ti.<\/p>\n<p><b class=\"spoiler_title\">U\u017eito\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\/sk\/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=\"sk_SK\" \/>\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\/sk\/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\u00fdkonnosti virtu\u00e1lnej m\u00e1quinas v VMware vSphere. \u010cas\u0165 1: CPU | ProHoster","description":"","canonical_url":"https:\/\/prohoster.info\/sk\/blog\/administrirovanie\/analiz-proizvoditelnosti-virtualnoj-mashiny-v-vmware-vsphere-chast-1-cpu","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"sk_SK","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\/sk\/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\/sk\/wp-json\/wp\/v2\/posts\/34335","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/sk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/sk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sk\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sk\/wp-json\/wp\/v2\/comments?post=34335"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/sk\/wp-json\/wp\/v2\/posts\/34335\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sk\/wp-json\/wp\/v2\/media\/25892"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/sk\/wp-json\/wp\/v2\/media?parent=34335"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/sk\/wp-json\/wp\/v2\/categories?post=34335"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/sk\/wp-json\/wp\/v2\/tags?post=34335"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}