Kubernetes yeML Workloads Quiz

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

1. Nderipi basa rekutanga reKubernetes scheduler yeML mabasa?
   • Kunyora kodhi yekudzidzira ye data masayendisiti
   • Kusarudza kuti ndeipi node musumbu inomhanya yega yega yemukati pod
   • Kunyora data rekudzidziswa otomatiki
   • Kusarudza iyo neural network architecture
2. Kubernetes anowanzoita sei kuti maGPU awanikwe kune pod?
   • Kuburikidza nemidziyo plugins inoshambadza zviwanikwa senge nvidia.com/gpu
   • Nekutumira email kune GPU mutengesi kune rezinesi
   • GPUs haigone kushandiswa muKubernetes
   • Nekushandura CPU cores kuita maGPU panguva yekumhanya
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. Ndechipi chishandiso chinowedzera ML-chaiyo hunyanzvi sevakagovera-vanodzidzisa vanoshanda pamusoro peKubernetes?
   • Microsoft Shoko
   • Apache yekushandira static mapeji ewebhu
   • A spreadsheet macro
   • Kubeflow
5. Nei Kubernetes yakanyatsokodzera kune kuputika kweML mabasa?
   • Inoronga mabasa pane zviwanikwa zvemahara uye inosunungura iyo Hardware kana mabasa apera
   • Iyo inochengetedza zvachose maGPU ese kune mumwe mushandisi
   • Inodzivirira chero basa ripi zvaro kubva pakukwira pasi
   • Inongomhanyisa chigaba chimwe pa cluster
6. Ko 'kuyera magumo ekupedzisira kusvika zero' zvinorevei paKubernetes?
   • Kudzima modhi zvachose
   • Kuisa chokwadi kusvika pazero muzana
   • Kumhanyisa zero makopi kana pasina traffic, wobva wadzvanya pachikumbiro chinotevera
   • Kuderedza ma parameter emuenzaniso kusvika zero
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