Explain Considerations of Model-Driven Telemetry (including Data Consumption and Data Storage) 350-901 Practice Questions (Page 1)

Questions 1–5

1. 1application · medium

   A network automation team wants to automatically reconfigure a router when a specific telemetry condition is detected, such as high CPU utilization for a sustained period. They have a telemetry pipeline that streams data to a collector. Which approach should they use to trigger the automated action?

   • AManually review the telemetry data in a dashboard and reconfigure the router when high CPU is observed.
   • BConfigure the router to send an email alert when CPU is high, and have an engineer manually reconfigure it.
   • CUse a stream processing platform (e.g., Apache Kafka Streams) to analyze the telemetry data in real-time and trigger a webhook to a configuration management system.
   • DStore the telemetry data in a time-series database and run a periodic batch job every hour to check for high CPU and trigger a reconfiguration.
   Reveal Solution Discussion Select an answer first
2. 2foundation · easy

   What is a common approach for processing telemetry data to detect anomalies?

   • AStoring all data in a relational database for manual SQL queries.
   • BUsing a message broker to permanently store data for future analysis.
   • CConfiguring devices to only send data when an anomaly is detected.
   • DApplying machine learning algorithms to identify patterns that deviate from normal behavior.
   Reveal Solution Discussion Select an answer first
4. 3foundation · easy

   Which scenario best illustrates a limitation of traditional polling-based monitoring that model-driven telemetry addresses?

   • AA network administrator must manually run CLI commands on each device to check interface status.
   • BA device sends an SNMP trap when an interface goes down, but the trap is lost due to network congestion.
   • CA monitoring system polls a device every 5 minutes and may miss transient spikes in CPU usage.
   • DA network engineer uses SNMP to retrieve interface counters on demand.
   Reveal Solution Discussion Select an answer first
5. 4expert · hard

   A large enterprise is migrating from SNMP polling to model-driven telemetry. They have a mix of legacy devices that only support SNMP and newer devices that support gRPC telemetry. They want a unified monitoring solution that can handle both types of devices without overloading the network. Which strategy should they adopt?

   • AReplace all legacy devices with newer ones that support gRPC telemetry.
   • BContinue using SNMP polling for all devices, but increase the polling frequency to match telemetry cadence.
   • CUse gRPC telemetry for all devices, even if they don't support it, by using a gateway that converts SNMP to gRPC.
   • DImplement a hybrid approach: use SNMP polling for legacy devices and gRPC telemetry for newer devices, with a collector that normalizes the data into a common format.
   Reveal Solution Discussion Select an answer first
6. 5foundation · easy

   Why are time-series databases well-suited for storing telemetry data?

   • AThey require no retention policies because they store data indefinitely.
   • BThey automatically encrypt all stored data.
   • CThey are optimized for timestamped data and efficient time-based queries.
   • DThey store data in a relational format that is easy to join with other tables.
   Reveal Solution Discussion Select an answer first