Kubernetes ee ML Workloads Quiz

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

1. Waa maxay shaqada aasaasiga ah ee jadwalka Kubernetes ee culeyska shaqada ML?
   • Qoritaanka xeerka tababarka ee saynisyahannada xogta
   • Go'aaminta noodhka kooxdu waxa ay waddaa qayb kasta oo weelaysan
   • Calaamadaynta xogta tababarka si toos ah
   • Doorashada qaab dhismeedka shabakada neerfaha
2. Sidee ayuu Kubernetes caadi ahaan uga dhigaya GPU-yada mid la heli karo?
   • Iyada oo loo marayo plugins-yada qalabka xayeysiiya ilaha sida nvidia.com/gpu
   • Adigoo iimayl ugu diraya iibiyaha GPU shatiga
   • GPU-yada laguma isticmaali karo Kubernetes
   • Adigoo u beddelaya xudunta CPU ee GPU-yada wakhtiga runtime
3. With a NoSchedule taint on a GPU node, what does a matching pod toleration do?
   • Forces the pod onto that node regardless of capacity
   • Changes the model’s optimizer
   • Allows the pod to be considered for that node without guaranteeing placement
   • Encrypts every GPU tensor
4. Qalabkee ayaa ku daraya awoodaha gaarka ah ee ML sida hawl wadeenada tababarka qaybsan ee korka Kubernetes?
   • Microsoft Eray
   • Apache ee loogu adeegayo bogagga shabakadda ee taagan
   • Macro xaashida
   • Kubeflow
5. Waa maxay sababta Kubernetes uu ugu habboon yahay culeysyada shaqada ee ML ee dillaacay?
   • Waxay jadwal u sameysaa shaqooyinka ilaha bilaashka ah waxayna sii daysaa qalabkaas marka ay shaqadu dhamaato
   • Waxay si joogto ah u xafidaysaa dhammaan GPU-yada hal isticmaale
   • Waxay ka hortagtaa shaqo kasta inay weligeed hoos u dhacdo
   • Waxa ay wadaa hal weel oo keliya koox kasta
6. Waa maxay 'kala-soocidda dhibcaha dhamaadka eber' waxay uga dhigan tahay Kubernetes?
   • Tirtiridda qaabka si joogto ah
   • Dejinta saxnaanta boqolkiiba eber
   • Koobiyada eber ordaya marka aysan jirin wax taraafig ah, ka dibna kor u qaadaya codsiga xiga
   • Hoos u dhigista cabbirka moodeelka ilaa eber
7. One ML pod requests four GPUs, while two nodes each have only two available. Can ordinary scheduling combine those nodes for that one pod?
   • Yes; every pod automatically spans all cluster nodes
   • Yes; the scheduler converts CPU cores into the missing GPUs
   • Only if the pod name contains the word distributed
   • No; the pod must fit on one eligible node
8. An inference service scales to zero while idle. Which user-facing delay should its team test when traffic resumes?
   • The training-set labeling speed
   • Cold-start delay while a serving replica becomes ready
   • The file size of an unrelated notebook
   • The number of discarded vocabulary tokens