Regarding Streaming's topology, what is wrong in the following description?

Question

Options

• AA Topology is a directed acyclic graph (DAG) connected by a set of Spout components and Bolt
• BTopology will keep running until it is explicitly killed
• CBusiness logic is encapsulated in Topology
• DA Topology can only specify to start one Worker process

• A
  6% (2)
• B
  3% (1)
• C
  17% (6)
• D
  75% (27)

Explanation

Option D is incorrect because Apache Storm topologies are explicitly designed to run across multiple Worker processes - you configure the number of workers via TOPOLOGY_WORKERS, allowing the topology to distribute computation across many JVMs or machines for parallelism and fault tolerance. Limiting a topology to a single worker would defeat the purpose of distributed stream processing.

Options A, B, and C are all accurate: a topology is indeed a DAG of Spouts (data sources) and Bolts (processors) (A); topologies are long-running by design and only stop when explicitly killed via the Storm UI or CLI (B); and the topology definition is precisely where business logic - the transformation and routing rules - lives (C).

Memory tip: Think of the "T" in Topology as standing for "Tunable workers" - the whole point of deploying to a cluster is scaling across many workers, so any claim that restricts it to one is a red flag.

No community discussion yet for this question.