A developer must deploy a containerized application for network device inventory management. The developer sets up a Kubernetes cluster on two separate hypervisors. The SLA is not currently meeting a specified maximum value for network latency/jitter. CPU/memory and disk I/O are functioning properly. Which two design approaches resolve the issue? (Choose two.)
To resolve issues with network latency and jitter in a Kubernetes cluster, upgrading the server NIC card and deploying the cluster to a bare metal server are effective approaches. Upgrading the NIC card can improve network performance by providing better throughput and lower latency. Deploying to a bare metal server eliminates the overhead associated with hypervisors, which can reduce latency and jitter, leading to a more stable and performant network infrastructure.
The answers are really confusing... A "pod" in kubernetes is an object, however here is considered the rack as a physical. If this is the case then BE are the options, if not, I am not sure... Just another confusing question of Cisco
To resolve issues with network latency and jitter in a Kubernetes cluster, upgrading the server NIC card (B) and deploying the cluster to a bare metal server (C) are effective approaches. These strategies directly target the network layer and the infrastructure, thereby reducing latency and jitter and improving overall network performance.
Correct answers: B & E https://netapp-trident.readthedocs.io/en/stable-v19.01/dag/kubernetes/kubernetes_cluster_architecture_considerations.html#cluster-
Choosing an architecture Regardless of the architecture that you choose, it’s important to understand the ramifications of high availability, scalability, and serviceability of the component services. Be sure to consider the effect on the applications being hosted by the Kubernetes or OpenShift cluster. The architecture of the storage infrastructure supporting the Kubernetes/OpenShift cluster and the hosted applications can also be affected by the chosen cluster architecture, such as where etcd is hosted. https://netapp-trident.readthedocs.io/en/stable-v19.01/dag/kubernetes/kubernetes_cluster_architecture_considerations.html#cluster-
I think B and E are correct> B) can improve the latency/jitter by increasing the NIC card (ex: from 10 to 100GB) D) They have to be spread in separate devices, but in the same location (pod) since they are a cluster.
Agree with the answers provided.
B - may improve network E - same pod/rack/switch also improves the network