{"id":39276,"date":"2019-10-31T22:29:10","date_gmt":"2019-10-31T19:29:10","guid":{"rendered":"https:\/\/prohoster.info\/blog\/proekt-open-data-hub-otkrytaya-platforma-mashinnogo-obucheniya-na-baze-red-hat-openshift\/"},"modified":"2019-10-31T22:29:10","modified_gmt":"2019-10-31T19:29:10","slug":"proekt-open-data-hub-otkrytaya-platforma-mashinnogo-obucheniya-na-baze-red-hat-openshift","status":"publish","type":"post","link":"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/proekt-open-data-hub-otkrytaya-platforma-mashinnogo-obucheniya-na-baze-red-hat-openshift","title":{"rendered":"Projekt Open Data Hub \u2013 otev\u0159en\u00e1 platforma pro strojov\u00e9 u\u010den\u00ed zalo\u017een\u00e1 na Red Hat OpenShift","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"<p>Budoucnost p\u0159i\u0161la, technologie um\u011bl\u00e9 inteligence a strojov\u00e9ho u\u010den\u00ed ji\u017e \u00fasp\u011b\u0161n\u011b pou\u017e\u00edvaj\u00ed va\u0161e obl\u00edben\u00e9 obchody, dopravn\u00ed spole\u010dnosti a dokonce i farmy chovaj\u00edc\u00ed kr\u016fty. <\/p>\n<p><img decoding=\"async\" alt=\"Projekt Open Data Hub \u2013 otev\u0159en\u00e1 platforma pro strojov\u00e9 u\u010den\u00ed zalo\u017een\u00e1 na Red Hat OpenShift\" src=\"\/wp-content\/uploads\/2019\/10\/ac77ba2d00012d4f3bee803c9c26bc9f.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n <br \/>\nA pokud n\u011bco existuje, znamen\u00e1 to, \u017ee na internetu o tom u\u017e n\u011bco je\u2026 otev\u0159en\u00fd projekt! Pod\u00edvejte se, jak Open Data Hub pom\u00e1h\u00e1 \u0161k\u00e1lovat nov\u00e9 technologie a vyh\u00fdbat se obt\u00ed\u017e\u00edm p\u0159i jejich zav\u00e1d\u011bn\u00ed.<br \/>\n<noindex><a rel=\"nofollow\" name=\"habracut\"><\/a><\/noindex><br \/>\nI p\u0159es v\u0161echny v\u00fdhody um\u011bl\u00e9 inteligence (artificial intelligence, AI) a strojov\u00e9ho u\u010den\u00ed (machine learning, ML) maj\u00ed organizace \u010dasto pot\u00ed\u017ee se \u0161k\u00e1lov\u00e1n\u00edm t\u011bchto technologi\u00ed. Hlavn\u00ed probl\u00e9my obvykle zahrnuj\u00ed n\u00e1sleduj\u00edc\u00ed:<\/p>\n<ul>\n<li><b>V\u00fdm\u011bna informac\u00ed a spolupr\u00e1ce<\/b> \u2013 sd\u00edlen\u00ed informac\u00ed bez zbyte\u010dn\u00fdch obt\u00ed\u017e\u00ed a spolupr\u00e1ce v re\u017eimu rychl\u00fdch iterac\u00ed je t\u00e9m\u011b\u0159 nemo\u017en\u00e9.<\/li>\n<li><b>P\u0159\u00edstup k dat\u016fm<\/b> \u2013 pro ka\u017edou \u00falohu je pot\u0159eba jej vytv\u00e1\u0159et znovu a ru\u010dn\u011b, co\u017e zab\u00edr\u00e1 spoustu \u010dasu.<\/li>\n<li><b>P\u0159\u00edstup na vy\u017e\u00e1d\u00e1n\u00ed<\/b> \u2013 nen\u00ed mo\u017en\u00e9 z\u00edskat on-demand p\u0159\u00edstup k n\u00e1stroj\u016fm a platform\u011b strojov\u00e9ho u\u010den\u00ed i k v\u00fdpo\u010detn\u00ed infrastruktu\u0159e.<\/li>\n<li><b>Produkce<\/b> \u2013 modely z\u016fst\u00e1vaj\u00ed ve f\u00e1zi prototypu a nejsou uvedeny do provozu.<\/li>\n<li><b>Sledov\u00e1n\u00ed a vysv\u011btlov\u00e1n\u00ed v\u00fdsledk\u016f pr\u00e1ce AI<\/b> \u2013 reprodukovatelnost, sledov\u00e1n\u00ed a vysv\u011btlov\u00e1n\u00ed v\u00fdsledk\u016f AI\/ML je obt\u00ed\u017en\u00e9.<\/li>\n<\/ul>\n<p>\nPokud z\u016fstanou tyto probl\u00e9my nevy\u0159e\u0161eny, negativn\u011b ovliv\u0148uj\u00ed rychlost, efektivitu a produktivitu cenn\u00fdch odborn\u00edk\u016f na zpracov\u00e1n\u00ed a anal\u00fdzu dat. To vede k jejich frustraci, zklam\u00e1n\u00ed v pr\u00e1ci a nakonec o\u010dek\u00e1v\u00e1n\u00ed podnik\u016f ohledn\u011b AI\/ML se nenapln\u00ed.<\/p>\n<p>Odpov\u011bdnost za \u0159e\u0161en\u00ed t\u011bchto probl\u00e9m\u016f le\u017e\u00ed na IT odborn\u00edc\u00edch, kte\u0159\u00ed mus\u00ed poskytnout datov\u00fdm analytik\u016fm \u2013 spr\u00e1vn\u011b, n\u011bco jako cloud. Pokud to roz\u0161\u00ed\u0159\u00edme, je t\u0159eba platforma, kter\u00e1 poskytuje svobodu volby a m\u00e1 snadn\u00fd, jednoduch\u00fd p\u0159\u00edstup. P\u0159itom by m\u011bla b\u00fdt rychl\u00e1, snadno p\u0159eprogramovateln\u00e1, \u0161k\u00e1lovateln\u00e1 na vy\u017e\u00e1d\u00e1n\u00ed a odoln\u00e1 v\u016f\u010di selh\u00e1n\u00ed. Vytv\u00e1\u0159en\u00ed takov\u00e9 platformy na z\u00e1klad\u011b technologi\u00ed s otev\u0159en\u00fdm zdrojov\u00fdm k\u00f3dem pom\u00e1h\u00e1 vyhnout se z\u00e1vislosti na dodavatel\u00edch a udr\u017eet dlouhodobou strategickou v\u00fdhodu v oblasti kontroly n\u00e1klad\u016f.<\/p>\n<p>P\u0159ed n\u011bkolika lety se n\u011bco podobn\u00e9ho d\u011blo ve v\u00fdvoji aplikac\u00ed a vedlo k vzniku mikroslu\u017eeb, hybridn\u00edch cloudov\u00fdch prost\u0159ed\u00ed, IT automatizace a agile proces\u016f. Aby se s t\u00edmto vyrovnali, IT specialist\u00e9 za\u010dali pou\u017e\u00edvat kontejnery, Kubernetes a otev\u0159en\u00e9 hybridn\u00ed cloudy.<\/p>\n<p>Nyn\u00ed se tato zku\u0161enost aplikuje na odpov\u011bdi na v\u00fdzvy AI. Proto IT specialist\u00e9 vytv\u00e1\u0159ej\u00ed platformy, kter\u00e9 se zakl\u00e1daj\u00ed na kontejnerech, umo\u017e\u0148uj\u00ed vytv\u00e1\u0159et AI\/ML slu\u017eby v r\u00e1mci agile proces\u016f, zrychluj\u00ed inovace a jsou konstruov\u00e1ny s ohledem na hybridn\u00ed cloud.<\/p>\n<p><img decoding=\"async\" alt=\"Projekt Open Data Hub \u2013 otev\u0159en\u00e1 platforma pro strojov\u00e9 u\u010den\u00ed zalo\u017een\u00e1 na Red Hat OpenShift\" src=\"\/wp-content\/uploads\/2019\/10\/a899093183092edd8b2902f47b0495a8.jpg\" style=\"display:block;margin: 0 auto;\" \/><\/p>\n<p>Budov\u00e1n\u00ed takov\u00e9 platformy za\u010dneme s Red Hat OpenShift, na\u0161\u00ed kontejnerovou Kubernetes platformou pro hybridn\u00ed cloud, kter\u00e1 m\u00e1 rychle rostouc\u00ed ekosyst\u00e9m softwarov\u00fdch a hardwarov\u00fdch ML \u0159e\u0161en\u00ed (NVIDIA, H2O.ai, Starburst, PerceptiLabs atd.). N\u011bkte\u0159\u00ed z klient\u016f Red Hat, jako nap\u0159\u00edklad BMW Group, ExxonMobil a dal\u0161\u00ed, ji\u017e nasadili kontejnerizovan\u00e9 \u0159et\u011bzce ML n\u00e1stroj\u016f a DevOps procesy zalo\u017een\u00e9 na t\u00e9to platform\u011b a jej\u00edm ekosyst\u00e9mu, aby uvedli sv\u00e9 ML architektury do pr\u016fmyslov\u00e9ho provozu a urychlili pr\u00e1ci datov\u00fdch analytik\u016f.<\/p>\n<p>Dal\u0161\u00edm d\u016fvodem, pro\u010d jsme spustili projekt Open Data Hub, je p\u0159edv\u00e9st p\u0159\u00edklad architektury na z\u00e1klad\u011b n\u011bkolika projekt\u016f se otev\u0159en\u00fdm k\u00f3dem a uk\u00e1zat, jak realizovat cel\u00fd \u017eivotn\u00ed cyklus ML \u0159e\u0161en\u00ed na platform\u011b OpenShift.<\/p>\n<h3>Projekt Open Data Hub<\/h3>\n<p>\nJe to projekt s otev\u0159en\u00fdm k\u00f3dem, kter\u00fd se vyv\u00edj\u00ed v r\u00e1mci p\u0159\u00edslu\u0161n\u00e9ho komunitn\u00edho v\u00fdvoje a realizuje pln\u00fd cyklus operac\u00ed \u2013 od na\u010d\u00edt\u00e1n\u00ed a transformace vstupn\u00edch dat a\u017e po tvorbu, tr\u00e9nov\u00e1n\u00ed a \u00fadr\u017ebu modelu \u2013 p\u0159i \u0159e\u0161en\u00ed \u00faloh AI\/ML pomoc\u00ed kontejner\u016f a Kubernetes na platform\u011b OpenShift. Tento projekt lze pova\u017eovat za referen\u010dn\u00ed implementaci, p\u0159\u00edklad toho, jak vybudovat otev\u0159en\u00e9 \u0159e\u0161en\u00ed typu \"AI\/ML jako slu\u017eba\" na z\u00e1klad\u011b OpenShift a odpov\u00eddaj\u00edc\u00edch n\u00e1stroj\u016f s otev\u0159en\u00fdm k\u00f3dem, jako jsou Tensorflow, JupyterHub, Spark a dal\u0161\u00ed. Je d\u016fle\u017eit\u00e9 poznamenat, \u017ee Red Hat tento projekt vyu\u017e\u00edv\u00e1 k poskytov\u00e1n\u00ed sv\u00fdch slu\u017eeb AI\/ML. Krom\u011b toho se OpenShift integruje s kl\u00ed\u010dov\u00fdmi softwarov\u00fdmi a hardwarov\u00fdmi ML \u0159e\u0161en\u00edmi od NVIDIA, Seldon, Starbust a dal\u0161\u00edch dodavatel\u016f, co\u017e usnad\u0148uje budov\u00e1n\u00ed a spou\u0161t\u011bn\u00ed vlastn\u00edch syst\u00e9m\u016f strojov\u00e9ho u\u010den\u00ed.<\/p>\n<p><img decoding=\"async\" alt=\"Projekt Open Data Hub \u2013 otev\u0159en\u00e1 platforma pro strojov\u00e9 u\u010den\u00ed zalo\u017een\u00e1 na Red Hat OpenShift\" src=\"\/wp-content\/uploads\/2019\/10\/85a4266250e1afca62bdc9406422510c.jpg\" style=\"display:block;margin: 0 auto;\" \/><\/p>\n<p>Projekt Open Data Hub se zam\u011b\u0159uje na n\u00e1sleduj\u00edc\u00ed kategorie u\u017eivatel\u016f a sc\u00e9n\u00e1\u0159e pou\u017eit\u00ed:<\/p>\n<ul>\n<li>Datov\u00fd analytik, kter\u00fd pot\u0159ebuje \u0159e\u0161en\u00ed pro implementaci ML projekt\u016f organizovan\u00e9 jako cloudov\u00e9 \u0159e\u0161en\u00ed se samoobslu\u017en\u00fdmi funkcemi.<\/li>\n<li>Datov\u00fd analytik, kter\u00fd pot\u0159ebuje maxim\u00e1ln\u00ed v\u00fdb\u011br z nejnov\u011bj\u0161\u00edch n\u00e1stroj\u016f a platforem AI\/ML s otev\u0159en\u00fdm zdrojov\u00fdm k\u00f3dem.<\/li>\n<li>Datov\u00fd analytik, kter\u00fd pot\u0159ebuje p\u0159\u00edstup k datov\u00fdm zdroj\u016fm p\u0159i tr\u00e9nov\u00e1n\u00ed model\u016f.<\/li>\n<li>Datov\u00fd analytik, kter\u00fd pot\u0159ebuje p\u0159\u00edstup k v\u00fdpo\u010detn\u00edm prost\u0159edk\u016fm (CPU, GPU, pam\u011b\u0165).<\/li>\n<li>Datov\u00fd analytik, kter\u00e9mu je t\u0159eba umo\u017enit spolupr\u00e1ci a v\u00fdm\u011bnu pracovn\u00edch v\u00fdsledk\u016f s kolegy, z\u00edsk\u00e1v\u00e1n\u00ed zp\u011btn\u00e9 vazby a implementaci vylep\u0161en\u00ed pomoc\u00ed rychl\u00fdch iterac\u00ed.<\/li>\n<li>Datov\u00fd analytik, kter\u00fd chce spolupracovat s v\u00fdvoj\u00e1\u0159i (a devops t\u00fdmy), aby jeho ML modely a v\u00fdsledky pracovn\u00edho procesu p\u0159e\u0161ly do produkce.<\/li>\n<li>Datov\u00fd in\u017een\u00fdr, kter\u00fd mus\u00ed zajistit, aby datov\u00ed analytici m\u011bli p\u0159\u00edstup k r\u016fzn\u00fdm datov\u00fdm zdroj\u016fm p\u0159i dodr\u017een\u00ed bezpe\u010dnostn\u00edch p\u0159edpis\u016f a po\u017eadavk\u016f.<\/li>\n<li>Administr\u00e1tor\/provozovatel IT syst\u00e9m\u016f, kter\u00fd pot\u0159ebuje snadno \u0159\u00eddit \u017eivotn\u00ed cyklus (instalace, konfigurace, aktualizace) komponent a technologi\u00ed s otev\u0159en\u00fdm zdrojov\u00fdm k\u00f3dem. Tak\u00e9 jsou pot\u0159eba odpov\u00eddaj\u00edc\u00ed n\u00e1stroje pro spr\u00e1vu a kv\u00f3tov\u00e1n\u00ed.<\/li>\n<\/ul>\n<p>\nProjekt Open Data Hub zahrnuje celou \u0159adu n\u00e1stroj\u016f s otev\u0159en\u00fdm zdrojov\u00fdm k\u00f3dem pro prov\u00e1d\u011bn\u00ed cel\u00e9ho cyklu operac\u00ed AI\/ML. Jako hlavn\u00ed pracovn\u00ed n\u00e1stroj datov\u00e9ho analytika se zde pou\u017e\u00edv\u00e1 Jupyter Notebook. Tento n\u00e1stroj je dnes \u0161iroce popul\u00e1rn\u00ed mezi odborn\u00edky na zpracov\u00e1n\u00ed a anal\u00fdzu dat, a Open Data Hub jim umo\u017e\u0148uje snadno vytv\u00e1\u0159et a spravovat pracovn\u00ed oblasti Jupyter Notebook\u016f pomoc\u00ed vestav\u011bn\u00e9ho JupyterHubu. Krom\u011b vytv\u00e1\u0159en\u00ed a importu Jupyter notebook\u016f obsahuje projekt Open Data Hub tak\u00e9 \u0159adu ji\u017e p\u0159ipraven\u00fdch notebook\u016f ve form\u011b knihovny AI Library.<\/p>\n<p>Tato knihovna p\u0159edstavuje sb\u00edrku open-source komponent strojov\u00e9ho u\u010den\u00ed a \u0159e\u0161en\u00ed pro standardn\u00ed sc\u00e9n\u00e1\u0159e, kter\u00e1 usnad\u0148uj\u00ed rychl\u00e9 prototypov\u00e1n\u00ed. JupyterHub je integrov\u00e1n s RBAC modelem p\u0159\u00edstupu OpenShift, co\u017e umo\u017e\u0148uje pou\u017e\u00edvat ji\u017e existuj\u00edc\u00ed \u00fa\u010dty OpenShift a implementovat jednotn\u00e9 p\u0159ihl\u00e1\u0161en\u00ed do syst\u00e9mu. Krom\u011b toho JupyterHub nab\u00edz\u00ed u\u017eivatelsky p\u0159\u00edv\u011btiv\u00e9 rozhran\u00ed nazvan\u00e9 spawner, pomoc\u00ed n\u011bho\u017e si u\u017eivatel m\u016f\u017ee snadno nakonfigurovat objem v\u00fdpo\u010detn\u00edch zdroj\u016f (procesorov\u00e1 j\u00e1dra, pam\u011b\u0165, GPU) pro vybran\u00fd Jupyter Notebook.<\/p>\n<p>A\u017e data analytik vytvo\u0159\u00ed a nakonfiguruje notebook, ve\u0161kerou dal\u0161\u00ed p\u00e9\u010di o n\u011bj p\u0159eb\u00edr\u00e1 pl\u00e1nova\u010d Kubernetes, kter\u00fd je sou\u010d\u00e1st\u00ed OpenShift. U\u017eivatel\u016fm zb\u00fdv\u00e1 pouze prov\u00e1d\u011bt sv\u00e9 experimenty, ukl\u00e1dat a sd\u00edlet v\u00fdsledky sv\u00e9 pr\u00e1ce. Krom\u011b toho mohou pokro\u010dil\u00ed u\u017eivatel\u00e9 p\u0159\u00edmo p\u0159istupovat k CLI shellu OpenShift p\u0159\u00edmo z Jupyter notebooks, aby vyu\u017eili Kubernetes primitiva, jako je Job, nebo funk\u010dnost OpenShift, nap\u0159\u00edklad Tekton nebo Knative. Nebo k tomu lze pou\u017e\u00edt pohodln\u00e9 u\u017eivatelsk\u00e9 rozhran\u00ed OpenShift, kter\u00e9 se naz\u00fdv\u00e1 \u201ewebov\u00e1 konzole OpenShift\u201c.<\/p>\n<p><img decoding=\"async\" alt=\"Projekt Open Data Hub \u2013 otev\u0159en\u00e1 platforma pro strojov\u00e9 u\u010den\u00ed zalo\u017een\u00e1 na Red Hat OpenShift\" src=\"\/wp-content\/uploads\/2019\/10\/601d63af1467b01df504954416effb3d.jpg\" style=\"display:block;margin: 0 auto;\" \/><\/p>\n<p><img decoding=\"async\" alt=\"Projekt Open Data Hub \u2013 otev\u0159en\u00e1 platforma pro strojov\u00e9 u\u010den\u00ed zalo\u017een\u00e1 na Red Hat OpenShift\" src=\"\/wp-content\/uploads\/2019\/10\/4647d8d979cc26b0c20cd3d63caed501.jpg\" style=\"display:block;margin: 0 auto;\" \/><\/p>\n<p>P\u0159i p\u0159echodu na dal\u0161\u00ed f\u00e1zi Open Data Hub umo\u017e\u0148uje spravovat datov\u00e9 kan\u00e1ly (data pipelines). K tomu se pou\u017e\u00edv\u00e1 objekt Ceph, kter\u00fd je poskytov\u00e1n jako S3-kompatibiln\u00ed objektov\u00fd \u00falo\u017ei\u0161t\u011b dat. Apache Spark zaji\u0161\u0165uje streaming dat z extern\u00edch zdroj\u016f nebo integrovan\u00e9ho \u00falo\u017ei\u0161t\u011b Ceph S3 a umo\u017e\u0148uje prov\u00e1d\u011bt p\u0159edb\u011b\u017en\u00e9 transformace dat. Apache Kafka zaji\u0161\u0165uje pokro\u010dilou spr\u00e1vu datov\u00fdch kan\u00e1l\u016f (kde lze prov\u00e1d\u011bt opakovan\u00e9 na\u010d\u00edt\u00e1n\u00ed a tak\u00e9 operace transformace, anal\u00fdzy a ukl\u00e1d\u00e1n\u00ed dat).<\/p>\n<p>Tak\u017ee datov\u00fd analytik z\u00edskal p\u0159\u00edstup k dat\u016fm a vybudoval model. Nyn\u00ed m\u00e1 touhu sd\u00edlet z\u00edskan\u00e9 v\u00fdsledky se sv\u00fdmi kolegy nebo v\u00fdvoj\u00e1\u0159i aplikac\u00ed a poskytnout jim sv\u016fj model na principu slu\u017eby. K tomu je pot\u0159eba v\u00fdstupn\u00ed server a Open Data Hub m\u00e1 takov\u00fd server, kter\u00fd se naz\u00fdv\u00e1 Seldon a umo\u017e\u0148uje publikovat model ve form\u011b RESTful slu\u017eby.<\/p>\n<p>V ur\u010dit\u00e9m okam\u017eiku se na serveru Seldon objevuje n\u011bkolik takov\u00fdch model\u016f a vznik\u00e1 pot\u0159eba monitorovat, jak jsou pou\u017e\u00edv\u00e1ny. V t\u00e9to souvislosti Open Data Hub nab\u00edz\u00ed kolekci relevantn\u00edch metrik a reportovac\u00ed engine zalo\u017een\u00fd na \u0161iroce pou\u017e\u00edvan\u00fdch open-source n\u00e1stroj\u00edch pro monitorov\u00e1n\u00ed Prometheus a Grafana. V\u00fdsledkem je zp\u011btn\u00e1 vazba pro sledov\u00e1n\u00ed pou\u017e\u00edv\u00e1n\u00ed AI model\u016f, zejm\u00e9na v produk\u010dn\u00edm prost\u0159ed\u00ed.<\/p>\n<p><img decoding=\"async\" alt=\"Projekt Open Data Hub \u2013 otev\u0159en\u00e1 platforma pro strojov\u00e9 u\u010den\u00ed zalo\u017een\u00e1 na Red Hat OpenShift\" src=\"\/wp-content\/uploads\/2019\/10\/1c95bcd2656ea8376ccc0a91fa266897.jpg\" style=\"display:block;margin: 0 auto;\" \/><\/p>\n<p>Open Data Hub tak poskytuje cloudov\u00fd p\u0159\u00edstup po celou dobu \u017eivotn\u00edho cyklu AI\/ML operac\u00ed, za\u010d\u00ednaje p\u0159\u00edstupem a p\u0159\u00edpravou dat a kon\u010de tr\u00e9nov\u00e1n\u00edm a pr\u016fmyslov\u00fdm nasazen\u00edm modelu.<\/p>\n<h3>D\u00e1v\u00e1me v\u0161e dohromady<\/h3>\n<p>\nTe\u010f se objevuje ot\u00e1zka, jak to v\u0161e zorganizovat administr\u00e1torovi OpenShift. A zde p\u0159ich\u00e1z\u00ed na \u0159adu speci\u00e1ln\u00ed Kubernetes operator pro projekty Open Data Hub.<\/p>\n<p><img decoding=\"async\" alt=\"Projekt Open Data Hub \u2013 otev\u0159en\u00e1 platforma pro strojov\u00e9 u\u010den\u00ed zalo\u017een\u00e1 na Red Hat OpenShift\" src=\"\/wp-content\/uploads\/2019\/10\/4bcb406f6886227ef357cd0318767ca2.jpg\" style=\"display:block;margin: 0 auto;\" \/><\/p>\n<p>Tento operator spravuje instalaci, konfiguraci a \u017eivotn\u00ed cyklus projektu Open Data Hub, v\u010detn\u011b nasazen\u00ed v\u00fd\u0161e zm\u00edn\u011bn\u00fdch n\u00e1stroj\u016f jako JupyterHub, Ceph, Spark, Kafka, Seldon, Prometheus a Grafana. Projekt Open Data Hub lze naj\u00edt ve webov\u00e9 konzoli OpenShift v sekci komunitn\u00edch oper\u00e1tor\u016f. T\u00edmto zp\u016fsobem m\u016f\u017ee administr\u00e1tor OpenShift nastavit, \u017ee p\u0159\u00edslu\u0161n\u00e9 projekty OpenShift pat\u0159\u00ed do kategorie \u201eprojekt Open Data Hub\u201c. Tato operace se prov\u00e1d\u00ed jednou. Pot\u00e9 datov\u00fd analytik prost\u0159ednictv\u00edm webov\u00e9 konzole OpenShift vstoup\u00ed do sv\u00e9ho projektov\u00e9ho prostoru a vid\u00ed, \u017ee pro jeho projekty je nainstalov\u00e1n a dostupn\u00fd p\u0159\u00edslu\u0161n\u00fd Kubernetes operator. N\u00e1sledn\u011b vytvo\u0159\u00ed instanci projektu Open Data Hub jedn\u00edm kliknut\u00edm a okam\u017eit\u011b z\u00edsk\u00e1 p\u0159\u00edstup k v\u00fd\u0161e popsan\u00fdm n\u00e1stroj\u016fm. A to v\u0161e lze nastavit v re\u017eimu vysok\u00e9 dostupnosti a odolnosti v\u016f\u010di poruch\u00e1m.<\/p>\n<p><img decoding=\"async\" alt=\"Projekt Open Data Hub \u2013 otev\u0159en\u00e1 platforma pro strojov\u00e9 u\u010den\u00ed zalo\u017een\u00e1 na Red Hat OpenShift\" src=\"\/wp-content\/uploads\/2019\/10\/811a85229d0f3e54a96de1adb9aaeb24.jpg\" style=\"display:block;margin: 0 auto;\" \/><\/p>\n<p>Pokud chcete projekt Open Data Hub vyzkou\u0161et na vlastn\u00ed k\u016f\u017ei, za\u010dn\u011bte s <noindex><a rel=\"nofollow\" href=\"https:\/\/opendatahub.io\/docs\/getting-started\/quick-installation.html\">pokyny k instalaci a \u00favodn\u00edm tutori\u00e1lem<\/a><\/noindex>. Technick\u00e9 podrobnosti architektury Open Data Hub lze naj\u00edt <noindex><a rel=\"nofollow\" href=\"https:\/\/opendatahub.io\/docs\/architecture.html\">zde<\/a><\/noindex>, pl\u00e1ny rozvoje projektu \u2013 <noindex><a rel=\"nofollow\" href=\"https:\/\/opendatahub.io\/docs\/roadmap\/future.html\">zde<\/a><\/noindex>. V budoucnu se pl\u00e1nuje realizace dal\u0161\u00ed integrace s Kubeflow, \u0159e\u0161en\u00ed r\u016fzn\u00fdch probl\u00e9m\u016f s regulac\u00ed dat a bezpe\u010dnost\u00ed, a tak\u00e9 organizace integrace se syst\u00e9mov\u00fdmi pravidly na z\u00e1klad\u011b Drools a Optaplanner. Vyj\u00e1d\u0159it sv\u016fj n\u00e1zor a st\u00e1t se \u00fa\u010dastn\u00edkem projektu <noindex><a rel=\"nofollow\" href=\"https:\/\/opendatahub.io\/docs\/additional.html#community\">Open Data Hub<\/a><\/noindex> lze na str\u00e1nce <noindex><a rel=\"nofollow\" href=\"https:\/\/opendatahub.io\/docs\/additional.html#community\">komunity<\/a><\/noindex>. <\/p>\n<p>Shrnut\u00ed: v\u00e1\u017en\u00e9 probl\u00e9my se \u0161k\u00e1lov\u00e1n\u00edm br\u00e1n\u00ed organizac\u00edm pln\u011b vyu\u017e\u00edt potenci\u00e1l um\u011bl\u00e9 inteligence a strojov\u00e9ho u\u010den\u00ed. Red Hat OpenShift je dlouhodob\u011b a \u00fasp\u011b\u0161n\u011b pou\u017e\u00edv\u00e1n k \u0159e\u0161en\u00ed podobn\u00fdch probl\u00e9m\u016f v softwarov\u00e9m sektoru. Projekt Open Data Hub, realizovan\u00fd v r\u00e1mci komunity otev\u0159en\u00e9ho v\u00fdvoje, nab\u00edz\u00ed referen\u010dn\u00ed architekturu pro organizaci cel\u00e9ho cyklu operac\u00ed AI\/ML zalo\u017een\u00fdch na hybridn\u00edm cloudu OpenShift. M\u00e1me jasn\u00fd a promy\u0161len\u00fd pl\u00e1n rozvoje tohoto projektu a jsme v\u00e1\u017en\u011b odhodl\u00e1ni vytvo\u0159it kolem n\u011bj aktivn\u00ed a plodnou komunitu v\u00fdvoje otev\u0159en\u00fdch AI \u0159e\u0161en\u00ed na platform\u011b OpenShift.<br \/>\n<br \/>Zdroj: <a content=\"nofollow\" rel=\"nofollow\" href=\"https:\/\/habr.com\/ru\/company\/redhatrussia\/blog\/472836\/\">habr.com<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"<p>\u0411\u0443\u0434\u0443\u0449\u0435\u0435 \u043d\u0430\u0441\u0442\u0443\u043f\u0438\u043b\u043e, \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0438 \u0438\u0441\u043a\u0443\u0441\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u0433\u043e \u0438\u043d\u0442\u0435\u043b\u043b\u0435\u043a\u0442\u0430 \u0438 \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f \u0443\u0436\u0435 \u0443\u0441\u043f\u0435\u0448\u043d\u043e \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u044e\u0442 \u0432\u0430\u0448\u0438 \u043b\u044e\u0431\u0438\u043c\u044b\u0435 \u043c\u0430\u0433\u0430\u0437\u0438\u043d\u044b, \u0442\u0440\u0430\u043d\u0441\u043f\u043e\u0440\u0442\u043d\u044b\u0435 \u043a\u043e\u043c\u043f\u0430\u043d\u0438\u0438 \u0438 \u0434\u0430\u0436\u0435 \u0444\u0435\u0440\u043c\u044b, \u0432\u044b\u0440\u0430\u0449\u0438\u0432\u0430\u044e\u0449\u0438\u0435 \u0438\u043d\u0434\u0435\u0435\u043a. \u0410 \u0435\u0441\u043b\u0438 \u0447\u0442\u043e-\u0442\u043e \u0441\u0443\u0449\u0435\u0441\u0442\u0432\u0443\u0435\u0442, \u0437\u043d\u0430\u0447\u0438\u0442 \u0432 \u0438\u043d\u0442\u0435\u0440\u043d\u0435\u0442\u0435 \u043f\u0440\u043e \u044d\u0442\u043e \u0443\u0436\u0435 \u0435\u0441\u0442\u044c\u2026 \u043e\u0442\u043a\u0440\u044b\u0442\u044b\u0439 \u043f\u0440\u043e\u0435\u043a\u0442! \u0421\u043c\u043e\u0442\u0440\u0438\u0442\u0435 \u043a\u0430\u043a Open Data Hub \u043f\u043e\u043c\u043e\u0433\u0430\u0435\u0442 \u043c\u0430\u0441\u0448\u0442\u0430\u0431\u0438\u0440\u043e\u0432\u0430\u0442\u044c \u043d\u043e\u0432\u044b\u0435 \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0438 \u0438 \u0438\u0437\u0431\u0435\u0433\u0430\u0442\u044c \u0442\u0440\u0443\u0434\u043d\u043e\u0441\u0442\u0435\u0439 \u043f\u0440\u0438 \u0438\u0445 \u0432\u043d\u0435\u0434\u0440\u0435\u043d\u0438\u0438. \u041f\u0440\u0438 \u0432\u0441\u0435\u0445 \u043f\u043b\u044e\u0441\u0430\u0445 \u0438\u0441\u043a\u0443\u0441\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u0433\u043e \u0438\u043d\u0442\u0435\u043b\u043b\u0435\u043a\u0442\u0430 (artificial [&hellip;]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":39277,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[688],"tags":[],"class_list":["post-39276","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=\"description\" content=\"\u0411\u0443\u0434\u0443\u0449\u0435\u0435 \u043d\u0430\u0441\u0442\u0443\u043f\u0438\u043b\u043e, \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0438 \u0438\u0441\u043a\u0443\u0441\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u0433\u043e \u0438\u043d\u0442\u0435\u043b\u043b\u0435\u043a\u0442\u0430 \u0438 \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f \u0443\u0436\u0435 \u0443\u0441\u043f\u0435\u0448\u043d\u043e \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u044e\u0442 \u0432\u0430\u0448\u0438 \u043b\u044e\u0431\u0438\u043c\u044b\u0435 \u043c\u0430\u0433\u0430\u0437\u0438\u043d\u044b, \u0442\u0440\u0430\u043d\u0441\u043f\u043e\u0440\u0442\u043d\u044b\u0435 \u043a\u043e\u043c\u043f\u0430\u043d\u0438\u0438 \u0438 \u0434\u0430\u0436\u0435 \u0444\u0435\u0440\u043c\u044b, \u0432\u044b\u0440\u0430\u0449\u0438\u0432\u0430\u044e\u0449\u0438\u0435 \u0438\u043d\u0434\u0435\u0435\u043a.\" \/>\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\/proekt-open-data-hub-otkrytaya-platforma-mashinnogo-obucheniya-na-baze-red-hat-openshift\" \/>\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\u041f\u0440\u043e\u0435\u043a\u0442 Open Data Hub \u2013 \u043e\u0442\u043a\u0440\u044b\u0442\u0430\u044f \u043f\u043b\u0430\u0442\u0444\u043e\u0440\u043c\u0430 \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f \u043d\u0430 \u0431\u0430\u0437\u0435 Red Hat OpenShift | ProHoster\" \/>\n\t\t<meta property=\"og:description\" content=\"\u0411\u0443\u0434\u0443\u0449\u0435\u0435 \u043d\u0430\u0441\u0442\u0443\u043f\u0438\u043b\u043e, \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0438 \u0438\u0441\u043a\u0443\u0441\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u0433\u043e \u0438\u043d\u0442\u0435\u043b\u043b\u0435\u043a\u0442\u0430 \u0438 \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f \u0443\u0436\u0435 \u0443\u0441\u043f\u0435\u0448\u043d\u043e \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u044e\u0442 \u0432\u0430\u0448\u0438 \u043b\u044e\u0431\u0438\u043c\u044b\u0435 \u043c\u0430\u0433\u0430\u0437\u0438\u043d\u044b, \u0442\u0440\u0430\u043d\u0441\u043f\u043e\u0440\u0442\u043d\u044b\u0435 \u043a\u043e\u043c\u043f\u0430\u043d\u0438\u0438 \u0438 \u0434\u0430\u0436\u0435 \u0444\u0435\u0440\u043c\u044b, \u0432\u044b\u0440\u0430\u0449\u0438\u0432\u0430\u044e\u0449\u0438\u0435 \u0438\u043d\u0434\u0435\u0435\u043a.\" \/>\n\t\t<meta property=\"og:url\" content=\"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/proekt-open-data-hub-otkrytaya-platforma-mashinnogo-obucheniya-na-baze-red-hat-openshift\" \/>\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=\"2019-10-31T19:29:10+00:00\" \/>\n\t\t<meta property=\"article:modified_time\" content=\"2019-10-31T19:29:10+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\udd47 Projekt Open Data Hub \u2013 otev\u0159en\u00e1 platforma strojov\u00e9ho u\u010den\u00ed na b\u00e1zi Red Hat OpenShift | ProHoster","description":"Budoucnost p\u0159i\u0161la, technologie um\u011bl\u00e9 inteligence a strojov\u00e9ho u\u010den\u00ed ji\u017e \u00fasp\u011b\u0161n\u011b pou\u017e\u00edvaj\u00ed va\u0161e obl\u00edben\u00e9 obchody, dopravn\u00ed spole\u010dnosti a dokonce i farmy chovaj\u00edc\u00ed kr\u016fty.","canonical_url":"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/proekt-open-data-hub-otkrytaya-platforma-mashinnogo-obucheniya-na-baze-red-hat-openshift","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\u041f\u0440\u043e\u0435\u043a\u0442 Open Data Hub \u2013 \u043e\u0442\u043a\u0440\u044b\u0442\u0430\u044f \u043f\u043b\u0430\u0442\u0444\u043e\u0440\u043c\u0430 \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f \u043d\u0430 \u0431\u0430\u0437\u0435 Red Hat OpenShift | ProHoster","og:description":"\u0411\u0443\u0434\u0443\u0449\u0435\u0435 \u043d\u0430\u0441\u0442\u0443\u043f\u0438\u043b\u043e, \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0438 \u0438\u0441\u043a\u0443\u0441\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u0433\u043e \u0438\u043d\u0442\u0435\u043b\u043b\u0435\u043a\u0442\u0430 \u0438 \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f \u0443\u0436\u0435 \u0443\u0441\u043f\u0435\u0448\u043d\u043e \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u044e\u0442 \u0432\u0430\u0448\u0438 \u043b\u044e\u0431\u0438\u043c\u044b\u0435 \u043c\u0430\u0433\u0430\u0437\u0438\u043d\u044b, \u0442\u0440\u0430\u043d\u0441\u043f\u043e\u0440\u0442\u043d\u044b\u0435 \u043a\u043e\u043c\u043f\u0430\u043d\u0438\u0438 \u0438 \u0434\u0430\u0436\u0435 \u0444\u0435\u0440\u043c\u044b, \u0432\u044b\u0440\u0430\u0449\u0438\u0432\u0430\u044e\u0449\u0438\u0435 \u0438\u043d\u0434\u0435\u0435\u043a.","og:url":"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/proekt-open-data-hub-otkrytaya-platforma-mashinnogo-obucheniya-na-baze-red-hat-openshift","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":"2019-10-31T19:29:10+00:00","article:modified_time":"2019-10-31T19:29:10+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"39276","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":"2026-01-24 01:33:20","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2021-02-28 14:14:14","updated":"2026-01-24 01:33:20","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\/39276","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=39276"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts\/39276\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/media\/39277"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/media?parent=39276"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/categories?post=39276"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/tags?post=39276"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}