Flink & Streaming interview questions

1. A.a condition where a slow downstream operator throttles upstream operators by slowing record flow through the pipeline
2. B.a marker in a stream that asserts event time has progressed past a timestamp, used to trigger event-time windows and handle lateness
3. C.a Flink state pattern that replicates a low-volume control stream (such as rules) to all parallel instances so it can join against a high-volume keyed stream
4. D.a Flink sink that ties external commits to checkpoints via prepare and commit phases to achieve end-to-end exactly-once delivery

Answer + AI explanation with Pro

1. A.async I/O
2. B.Exactly-once vs at-least-once checkpointing
3. C.Watermark
4. D.Event time vs processing time

Answer + AI explanation with Pro

1. A.Watermark — a marker in a stream that asserts event time has progressed past a timestamp, used to trigger event-time windows and handle lateness
2. B.Watermark — the Flink model that processes each event individually for low latency, contrasted with Spark Structured Streaming's small batches
3. C.Watermark — the Complex Event Processing library that detects ordered patterns of events in a stream using a pattern API with conditions, quantifiers, and time constraints
4. D.Watermark — the Flink state backend storing keyed state in an embedded RocksDB instance on local disk, enabling state larger than memory with incremental checkpoints

Answer + AI explanation with Pro

1. A.the Flink feature decoupling checkpointing from state materialization by continuously logging state changes, reducing checkpoint duration and tail latency
2. B.a condition where a slow downstream operator throttles upstream operators by slowing record flow through the pipeline
3. C.the Flink 2.0 architecture that stores primary operator state on remote storage like S3 with local disk as cache for cloud-native elasticity
4. D.a periodic, consistent snapshot of all operator state in Flink that enables exactly-once recovery after a failure

Answer + AI explanation with Pro

1. A.exactly-once sink
2. B.RocksDB state backend
3. C.Checkpoint
4. D.watermark

Answer + AI explanation with Pro

1. A.Checkpoint — a Flink sink that ties external commits to checkpoints via prepare and commit phases to achieve end-to-end exactly-once delivery
2. B.Checkpoint — a condition where a slow downstream operator throttles upstream operators by slowing record flow through the pipeline
3. C.Checkpoint — a periodic, consistent snapshot of all operator state in Flink that enables exactly-once recovery after a failure
4. D.Checkpoint — the special marker Flink injects into streams so all operators snapshot their state at a consistent point for fault-tolerant recovery

Answer + AI explanation with Pro

1. A.the unified Flink connector (KafkaSource) supporting bounded and unbounded reads with offset initialization, watermark generation, and exactly-once via committed offsets
2. B.Flink state partitioned by key so each parallel instance maintains independent state for its assigned keys
3. C.a manually triggered, durable snapshot of Flink state used for upgrades, rescaling, and version migration
4. D.the distinction between when an event actually occurred and when it is processed, the former giving deterministic results despite delays

Answer + AI explanation with Pro

1. A.Backpressure
2. B.changelog state backend
3. C.Savepoint
4. D.checkpoint barrier

Answer + AI explanation with Pro

1. A.Savepoint — a Flink state pattern that replicates a low-volume control stream (such as rules) to all parallel instances so it can join against a high-volume keyed stream
2. B.Savepoint — the special marker Flink injects into streams so all operators snapshot their state at a consistent point for fault-tolerant recovery
3. C.Savepoint — a manually triggered, durable snapshot of Flink state used for upgrades, rescaling, and version migration
4. D.Savepoint — the unified Flink connector (KafkaSource) supporting bounded and unbounded reads with offset initialization, watermark generation, and exactly-once via committed offsets

Answer + AI explanation with Pro

1. A.a Flink mechanism using an OutputTag to emit selected records (such as late events or errors) to a secondary stream separate from the main output
2. B.the Flink feature decoupling checkpointing from state materialization by continuously logging state changes, reducing checkpoint duration and tail latency
3. C.a Flink state pattern that replicates a low-volume control stream (such as rules) to all parallel instances so it can join against a high-volume keyed stream
4. D.the distinction between when an event actually occurred and when it is processed, the former giving deterministic results despite delays

Answer + AI explanation with Pro

1. A.True streaming vs micro-batch
2. B.Event time vs processing time
3. C.Keyed state
4. D.Stateful stream processing

Answer + AI explanation with Pro

1. A.Event time vs processing time — a Flink mechanism using an OutputTag to emit selected records (such as late events or errors) to a secondary stream separate from the main output
2. B.Event time vs processing time — computation that retains information across events, such as running aggregates, kept in fault-tolerant operator state
3. C.Event time vs processing time — Flink state partitioned by key so each parallel instance maintains independent state for its assigned keys
4. D.Event time vs processing time — the distinction between when an event actually occurred and when it is processed, the former giving deterministic results despite delays

Answer + AI explanation with Pro

1. A.Flink state partitioned by key so each parallel instance maintains independent state for its assigned keys
2. B.a periodic, consistent snapshot of all operator state in Flink that enables exactly-once recovery after a failure
3. C.the Flink model that processes each event individually for low latency, contrasted with Spark Structured Streaming's small batches
4. D.computation that retains information across events, such as running aggregates, kept in fault-tolerant operator state

Answer + AI explanation with Pro

1. A.Kafka source
2. B.Backpressure
3. C.Keyed state
4. D.Event time vs processing time

Answer + AI explanation with Pro

1. A.Keyed state — the choice between aligned barriers that deduplicate state versus faster unaligned or relaxed modes that may reprocess
2. B.Keyed state — a Flink marker flowing with the stream that asserts no events earlier than its timestamp should still arrive, driving event-time window firing
3. C.Keyed state — Flink state partitioned by key so each parallel instance maintains independent state for its assigned keys
4. D.Keyed state — the Flink 2.0 disaggregated state store built for asynchronous remote access and lightweight checkpoint and recovery

Answer + AI explanation with Pro

1. A.a manually triggered, durable snapshot of Flink state used for upgrades, rescaling, and version migration
2. B.a Flink sink that ties external commits to checkpoints via prepare and commit phases to achieve end-to-end exactly-once delivery
3. C.the Flink feature decoupling checkpointing from state materialization by continuously logging state changes, reducing checkpoint duration and tail latency
4. D.the Flink state backend storing keyed state in an embedded RocksDB instance on local disk, enabling state larger than memory with incremental checkpoints

Answer + AI explanation with Pro

1. A.Two-phase commit sink
2. B.Checkpoint
3. C.Backpressure
4. D.side output

Answer + AI explanation with Pro

1. A.Two-phase commit sink — the special marker Flink injects into streams so all operators snapshot their state at a consistent point for fault-tolerant recovery
2. B.Two-phase commit sink — a Flink sink that ties external commits to checkpoints via prepare and commit phases to achieve end-to-end exactly-once delivery
3. C.Two-phase commit sink — the Flink operator that issues concurrent requests to external systems and processes responses out of order, hiding lookup latency without blocking the pipeline
4. D.Two-phase commit sink — a Flink marker flowing with the stream that asserts no events earlier than its timestamp should still arrive, driving event-time window firing

Answer + AI explanation with Pro

1. A.Flink state partitioned by key so each parallel instance maintains independent state for its assigned keys
2. B.the Complex Event Processing library that detects ordered patterns of events in a stream using a pattern API with conditions, quantifiers, and time constraints
3. C.the Flink 2.0 architecture that stores primary operator state on remote storage like S3 with local disk as cache for cloud-native elasticity
4. D.a Flink sink achieving end-to-end exactly-once delivery via a two-phase commit that aligns external commits with Flink checkpoints

Answer + AI explanation with Pro

1. A.ForSt state backend
2. B.Disaggregated state
3. C.changelog state backend
4. D.checkpoint barrier

Answer + AI explanation with Pro

1. A.Disaggregated state — a manually triggered, durable snapshot of Flink state used for upgrades, rescaling, and version migration
2. B.Disaggregated state — a Flink marker flowing with the stream that asserts no events earlier than its timestamp should still arrive, driving event-time window firing
3. C.Disaggregated state — a Flink state pattern that replicates a low-volume control stream (such as rules) to all parallel instances so it can join against a high-volume keyed stream
4. D.Disaggregated state — the Flink 2.0 architecture that stores primary operator state on remote storage like S3 with local disk as cache for cloud-native elasticity

Answer + AI explanation with Pro

1. A.the choice between aligned barriers that deduplicate state versus faster unaligned or relaxed modes that may reprocess
2. B.the Flink 2.0 disaggregated state store built for asynchronous remote access and lightweight checkpoint and recovery
3. C.Flink state partitioned by key so each parallel instance maintains independent state for its assigned keys
4. D.a Flink sink that ties external commits to checkpoints via prepare and commit phases to achieve end-to-end exactly-once delivery

Answer + AI explanation with Pro

1. A.Stateful stream processing
2. B.ForSt state backend
3. C.Flink CEP
4. D.Watermark

Answer + AI explanation with Pro

1. A.ForSt state backend — Flink state partitioned by key so each parallel instance maintains independent state for its assigned keys
2. B.ForSt state backend — the Flink feature decoupling checkpointing from state materialization by continuously logging state changes, reducing checkpoint duration and tail latency
3. C.ForSt state backend — a Flink mechanism using an OutputTag to emit selected records (such as late events or errors) to a secondary stream separate from the main output
4. D.ForSt state backend — the Flink 2.0 disaggregated state store built for asynchronous remote access and lightweight checkpoint and recovery

Answer + AI explanation with Pro

1. A.the Flink model that processes each event individually for low latency, contrasted with Spark Structured Streaming's small batches
2. B.a manually triggered, durable snapshot of Flink state used for upgrades, rescaling, and version migration
3. C.a Flink sink achieving end-to-end exactly-once delivery via a two-phase commit that aligns external commits with Flink checkpoints
4. D.Flink state partitioned by key so each parallel instance maintains independent state for its assigned keys

Answer + AI explanation with Pro

1. A.True streaming vs micro-batch
2. B.changelog state backend
3. C.Exactly-once vs at-least-once checkpointing
4. D.checkpoint barrier

Answer + AI explanation with Pro

1. A.True streaming vs micro-batch — computation that retains information across events, such as running aggregates, kept in fault-tolerant operator state
2. B.True streaming vs micro-batch — the Flink model that processes each event individually for low latency, contrasted with Spark Structured Streaming's small batches
3. C.True streaming vs micro-batch — the Complex Event Processing library that detects ordered patterns of events in a stream using a pattern API with conditions, quantifiers, and time constraints
4. D.True streaming vs micro-batch — a manually triggered, durable snapshot of Flink state used for upgrades, rescaling, and version migration

Answer + AI explanation with Pro

1. A.a condition where a slow downstream operator throttles upstream operators by slowing record flow through the pipeline
2. B.a Flink sink that ties external commits to checkpoints via prepare and commit phases to achieve end-to-end exactly-once delivery
3. C.the special marker Flink injects into streams so all operators snapshot their state at a consistent point for fault-tolerant recovery
4. D.the Flink state backend storing keyed state in an embedded RocksDB instance on local disk, enabling state larger than memory with incremental checkpoints

Answer + AI explanation with Pro

1. A.Savepoint
2. B.Event time vs processing time
3. C.Backpressure
4. D.Checkpoint

Answer + AI explanation with Pro

1. A.Backpressure — Flink state partitioned by key so each parallel instance maintains independent state for its assigned keys
2. B.Backpressure — the special marker Flink injects into streams so all operators snapshot their state at a consistent point for fault-tolerant recovery
3. C.Backpressure — a condition where a slow downstream operator throttles upstream operators by slowing record flow through the pipeline
4. D.Backpressure — a manually triggered, durable snapshot of Flink state used for upgrades, rescaling, and version migration

Answer + AI explanation with Pro