Okwu mmalite
Mmụba ngwa ngwa nke nkà na ụzụ emeela ka ọganihu dị egwu na mgbagwoju anya nke usoro IT ọgbara ọhụrụ. N'ihi ya, ndị otu IT na-eche ihe ịma aka ndị a na-enwetụbeghị ụdị ya ihu na ijikwa na idowe usoro ndị a, na-eduga n'ịbawanye oge nkwụsịtụ, mbelata arụpụtaghị ihe, na ọnụ ahịa ọrụ dị elu. Iji dozie ihe ịma aka ndị a, echiche nke Artificial Intelligence for IT Operations (AIOps) apụtala dị ka ihe na-agbanwe egwuregwu. AIOps na-akwalite mmụta igwe yana AI iji gbanwee usoro ọrụ DevOps, na-eme ka njikwa IT na-arụ ọrụ, nyocha amụma, na akpaaka.
N'isiokwu a, anyị ga-enyocha ọdịnihu AIOps, uru ya na atụmatụ mmejuputa. Anyị ga-abanyekwa n'ime ọmụmụ ihe gbasara ụwa ma nye ngọngọ koodu mgbagwoju anya iji gosipụta ngwa AIOps na njikwa IT.
Gịnị bụ AIops?
AIOps bụ mpaghara na-apụta nke na-ejikọta ike ọgụgụ isi (AI) na mmụta igwe (ML) na arụmọrụ IT (ITOps) iji melite arụmọrụ, ịdị irè na ntụkwasị obi nke ọrụ IT. nyiwe AIOps na-eji nyocha data, ihe nlegharị anya amụma, na akpaaka iji chọpụta na dozie okwu IT tupu ha aghọọ nnukwu nsogbu.
Enwere ike itinye AIops na mpaghara dị iche iche nke njikwa IT, gụnyere:
- ** Ntụle amụma ***: nyiwe AIOps na-eji igwe mmụta algọridim iji nyochaa data akụkọ ihe mere eme ma buru amụma ihe nwere ike ime IT, na-eme ka mmezi na-arụ ọrụ na-ebelata oge.
- ** Nlekọta ihe mberede ***: AIOps nwere ike ịchọpụta ozugbo ma dozie ihe omume IT, na-ebelata oge iji chọpụta (MTTD) na pụtara oge iji dozie (MTTR).
- **Automation ***: AIOps nwere ike megharịa ọrụ oge niile, dị ka nhazi nhazi, njikwa patch, na nkwado ndabere na mpaghara mgbake.
Uru AIOps
Uru nke AIOps dị ọtụtụ ma dị ịrịba ama:
- **Emere arụmọrụ ***: AIOps na-arụ ọrụ oge niile, na-ahapụ ndị ọrụ IT ka ha lekwasị anya na atụmatụ atụmatụ.
- ** Enwekwukwa ntụkwasị obi ***: AIOps na-ebu amụma ma na-egbochi nsogbu IT, na-ebelata oge nkwụsịtụ na imeziwanye ntụkwasị obi usoro n'ozuzu ya.
- ** Mmụba Mmepụta ***: AIOps na-enyere ndị otu IT aka ịzaghachi ngwa ngwa na ihe omume, na-ebelata oge iji chọpụta (MTTD) na pụtara oge iji dozie (MTTR).
- ** Ime mkpebi ka mma ***: AIOps na-enye ndị otu IT nghọta oge na nyocha amụma, na-eme ka mkpebi data na-eduzi.
Usoro mmejuputa
Ime AIOps chọrọ usoro atụmatụ:
- ** Nchịkọta data ***: Anakọta data IT site na isi mmalite dị iche iche, gụnyere ndekọ, metrics, na faịlụ nhazi.
- ** Data Integration ***: Jikọta data sitere na isi mmalite dị iche iche n'ime otu ikpo okwu.
- ** Mmụta igwe ***: Tinye algọridim mmụta igwe iji nyochaa data ma buru amụma okwu IT nwere ike.
- ** akpaaka ***: Megharịa ọrụ oge niile na usoro ọrụ.
- ** Nleba anya na nzaghachi ***: Na-aga n'ihu na-enyocha arụmọrụ AIOps ma nye nzaghachi iji melite izi ezi na ịdị irè.
Ọmụmụ ihe gbasara ụwa n'ezie
Ọtụtụ ụlọ ọrụ emejuputala AIOps nke ọma iji gbanwee usoro njikwa IT ha:
- ** Verizon ***: Verizon mejuputa AIOps iji melite ntụkwasị obi netwọk ma belata oge nkwụsị. Igwe ikpo okwu jiri mmụta igwe na-ebu amụma ma gbochie nsogbu netwọkụ, na-ebute mbelata 30% na oge ọdịda.
- ** AT & T ***: AT & T mejuputa AIOps iji meziwanye njikwa ihe omume ma belata oge iji chọpụta (MTTD) na pụtara oge iji dozie (MTTR). Igwe ikpo okwu jiri akpaaka iji chọpụta ma dozie ihe omume IT, na-ebute mbelata 40% na MTTD yana mbelata 50% na MTTR.
- **Microsoft ***: Microsoft mebere AIOps iji kwalite amụma amụma na akpaaka. Ikpo okwu jiri mmụta igwe na-ebu amụma ihe nwere ike ime IT, na-enyere aka ịrụ ọrụ nke ọma yana ibelata oge ọdịda.
Usoro ngọngọ: AIOps Ihe Nlereanya na Python
Nke a bụ ngọngọ koodu mgbagwoju anya na Python nke gosipụtara ngwa AIOps na njikwa IT: "Python mbubata pandas ka pd si sklearn.ensemble mbubata RandomForestClassifier si sklearn.model_selection mbubata train_test_split si sklearn.metrics mbubata ziri ezi_score
Budata data
data = pd.read_csv ('it_data.csv')
Hazie data
X = data.dobe (['omume'], axis=1) y = data['omume']
Kewaa data n'ime usoro ọzụzụ na nnwale
X_train, X_test, y_train, y_test = train_test_split (X, y, test_size=0.2, random_state=42)
Zụlite classifier ọhịa random
rfc = RandomForestClassifier (n_estimators=100, random_state=42) rfc.fit (X_train, y_train)
Kwuo amụma
y_pred = rfc.predict(X_test)
Nyochaa ihe nlereanya
izi ezi = izi ezi_score (y_test, y_pred) ebipụta (f'Ezizi: {izizi:.2f}')
Jiri ihe atụ buru amụma okwu IT nwere ike ime
new_data = pd.DataFrame ({'cpu_usage': [80], 'memory_usage': [90]}) amụma = rfc.predict(new_data) ọ bụrụ amụma [0] == 1: ebipụta ('Achọpụtara okwu IT nwere ike ime!') `` Ihe ngọngọ koodu a na-eji ihe nhazi oke ọhịa na-enweghị usoro iji buru amụma ihe nwere ike ime IT dabere na CPU na data ojiji ebe nchekwa. A zụrụ ihe nlereanya ahụ na data akụkọ ihe mere eme ma jiri ya mee amụma na data ọhụrụ.
Nlere anya: AIOps Platform Architecture
Nke a bụ nleba anya nke ihe owuwu ikpo okwu AIOps na-eji Chart.js:
chaatị { "ụdị": "pieChart", "data": { "labels": ["Nchịkọta data", "Mmekọrịta data", "Mmụta igwe", "Automation", "Nleba anya na nzaghachi"], "datasets": [ { "label": "AIOps Platform Architecture", "data": [30, 20, 20, 15, 15], "backgroundColor": ["#FF6384", "#36A2EB", "#FFCE56", "#4BC0C0", "#FF99CC"], "hoverBackgroundColor": ["#FF99CC", "#36A2EB", "#FFCE56", "#4BC0C0", "#FF99CC"] } ] }, "Nhọrọ": { " aha": { "ngosi": eziokwu, "text": "AIOps Platform Architecture" }, "akụkọ": { "ngosi": eziokwu } } }
Ọhụụ a na-egosi akụkụ dị iche iche nke ihe owuwu ikpo okwu AIOps, gụnyere nchịkọta data, ntinye data, mmụta igwe, akpaaka, na nleba anya na nzaghachi.
Mmechi
AIOps bụ teknụzụ na-agbanwe egwuregwu nke na-agbanwe ka ndị otu IT na-ejikwa ma na-ejigide sistemụ IT dị mgbagwoju anya. Site n'ịkwalite mmụta igwe na AI, AIOps na-enyere njikwa IT na-arụsi ọrụ ike, nyocha amụma, na akpaaka, na-eduga n'ịrụ ọrụ nke ọma, ntụkwasị obi na-abawanye, yana mmụba arụpụtaghị ihe. Site na itinye AIOps n'ọrụ, otu nwere ike ibelata oge nkwụsịtụ, melite ime mkpebi, ma nweta oke asọmpi n'ahịa.
Ka teknụzụ na-aga n'ihu na-etolite, anyị nwere ike ịtụ anya ịhụ ọbụna ngwa ọhụrụ nke AIOps na njikwa IT. Site n'ikike ya ịgbanwe ụzọ anyị si ejikwa sistemụ IT, AIOps bụ ubi na-atọ ụtọ ma na-eto ngwa ngwa nke kwesịrị ileba anya.