Streaming Systems

Streaming Systems

Kafka, Flink, Pulsar, exactly-once, CDC, tiered storage, backpressure.

54Articles
54Topics covered
Articles in this category

All 54 articles, sorted alphabetically

Advertisement
ARTICLE · 01

Change Data Capture in 2026

Debezium Decodable native CDC.

Read article
ARTICLE · 02

Dead Letter Queues

Pattern retention and re-processing.

Read article
ARTICLE · 03

Debezium CDC to Kafka

Capturing Postgres MySQL and MongoDB changes as event streams.

Read article
ARTICLE · 04

Event Sourcing vs Event Streaming

When the same word means different architectures.

Read article
ARTICLE · 05

Exactly-Once Semantics Across Systems

What actually gets guaranteed and where the gaps are.

Read article
ARTICLE · 06

Exactly-Once Semantics in Streaming

What 'exactly-once' actually means in Kafka, Flink, and Spark Structured Streaming.

Read article
ARTICLE · 07

Flink Checkpoint Design

Frequency, alignment, and what gets lost.

Read article
ARTICLE · 08

Flink State Backends Compared

RocksDB vs in-memory and when each one fails.

Read article
ARTICLE · 09

Kafka KRaft vs ZooKeeper

Migrating to the new Raft-based metadata quorum.

Read article
ARTICLE · 10

Kafka Partitioning Strategies

Pick the key wrong, regret it later.

Read article
ARTICLE · 11

Kafka Tiered Storage

S3-backed cold tier in production.

Read article
ARTICLE · 12

Kafka vs Pulsar vs Redpanda

Architecture differences between the three top streaming brokers + when each one wins.

Read article
ARTICLE · 13

Materialize for Streaming SQL

Incremental view maintenance over Kafka with Postgres-style queries.

Read article
ARTICLE · 14

Redpanda Tiered Storage

Offloading cold data to S3 while keeping hot writes local.

Read article
ARTICLE · 15

Schema Registry Patterns

Avro Protobuf JSON Schema in 2026.

Read article
ARTICLE · 16

Apache Flink Overview

What Flink is, its stateful streaming model, and how it differs from Spark Streaming.

Read article
ARTICLE · 17

Flink State

How Flink manages state per-key and per-operator, backed by RocksDB for durability.

Read article
ARTICLE · 18

Flink Windowing

How Flink windowing groups events into time or count-based windows for aggregation.

Read article
ARTICLE · 19

Kafka Consumers and Consumer Groups

How consumer groups coordinate to distribute partition consumption, and how offsets track progress.

Read article
ARTICLE · 20

Kafka Exactly-Once Semantics

How Kafka achieves exactly-once processing via idempotent producers and transactional writes.

Read article
ARTICLE · 21

Apache Kafka Overview

What Kafka is, its append-only log design, and how it powers streaming and event-driven architectures.

Read article
ARTICLE · 22

Kafka Partitions

How partitions distribute topic data across brokers and enable parallel consumption.

Read article
ARTICLE · 23

AWS Kinesis

How Kinesis Data Streams provide Kafka-like functionality as a managed AWS service.

Read article
ARTICLE · 24

Apache Pulsar

What Pulsar is, how its architecture (compute-storage separation) differs from Kafka, and when to use it.

Read article
ARTICLE · 25

Spark Streaming

How Spark Streaming (and Structured Streaming) process streams as micro-batches on the Spark engine.

Read article
ARTICLE · 26

Stream-Table Duality

Same data two views in Kafka Streams and Flink SQL.

Read article
ARTICLE · 27

Streaming watermark architecture

Deep-dive on streaming watermarks: event time, watermark, trigger, late data, allowed lateness, side output, retract.

Read article
ARTICLE · 28

Backpressure architecture

Deep-dive on backpressure in stream processing: why unbounded buffering causes stateful crashes, credit-based flow control between operator subtasks, …

Read article
ARTICLE · 29

Apache Beam architecture

Deep-dive on Apache Beam: PCollection, transforms, runners, windowing, triggers, watermark, state + timers, side inputs.

Read article
ARTICLE · 30

CDC architecture

Deep-dive on Change Data Capture: log-based capture (WAL/binlog), snapshot, schema evolution, delivery semantics, ordering, sinks.

Read article
ARTICLE · 31

Kafka consumer group rebalancing architecture

Deep-dive on Kafka consumer group rebalancing: the group coordinator and heartbeat liveness, revoke-and-assign phases, offset commits at the rebalance…

Read article
ARTICLE · 32

Exactly-once stream processing architecture

Deep-dive on end-to-end exactly-once semantics: replayable Kafka sources, Flink barrier-aligned snapshots, two-phase-commit sinks bound to checkpoint …

Read article
ARTICLE · 33

Apache Flink Architecture in Depth

A 2500-word walkthrough of Apache Flink architecture: JobManager, TaskManagers, dataflow graph, state backends, sources/sinks, windows, watermarks, ch…

Read article
ARTICLE · 34

Flink Async I/O architecture

Deep-dive on Flink Async I/O: overlapping many in-flight external lookups through a bounded-capacity buffer, the AsyncFunction and ResultFuture contra…

Read article
ARTICLE · 35

Flink savepoints architecture

Deep-dive on Apache Flink savepoints: user-triggered canonical-format state snapshots keyed by stable operator UIDs, exactly-once continuity across up…

Read article
ARTICLE · 36

Flink state architecture

Deep-dive on Flink state: keyed and operator state, backends, checkpoints, savepoints, barrier alignment, restore, rescale, TTL.

Read article
ARTICLE · 37

Apache Hudi architecture

Deep-dive on Apache Hudi: COW vs MOR, timeline, record index, base + log, compaction, clustering, cleaning, incremental read.

Read article
ARTICLE · 38

Iceberg architecture

Deep-dive on Apache Iceberg: data files, manifests, metadata, snapshots, partition specs, hidden partitioning, time travel, compaction.

Read article
ARTICLE · 39

Apache Kafka Streaming Architecture in Depth

A 2500-word walkthrough of Apache Kafka: producers, brokers with KRaft, topics + partitions, replication, schema registry, connect, stream processing,…

Read article
ARTICLE · 40

Kafka ISR replication architecture - leaders, followers, and the in-sync replica set

Deep-dive on Kafka's replication protocol: leader and follower fetch, the in-sync replica set, high watermark and commit sema…

Read article
ARTICLE · 41

Kafka ordering architecture

Deep-dive on Kafka ordering: partitioning, keys, idempotent producer, transactions, isolation, cooperative rebalance, and repartitioning.

Read article
ARTICLE · 42

Kafka client quotas architecture

Deep-dive on Kafka quotas: user/client-id entity precedence, produce/fetch/request/mutation quota types, the sliding-window rate sampler, the throttle…

Read article
ARTICLE · 43

Kafka Streams

Deep-dive on Kafka Streams: processor topologies over KStream/KTable, local RocksDB state stores backed by changelog topics, partitioning and stream t…

Read article
ARTICLE · 44

Kafka log compaction architecture

Deep-dive on Kafka log compaction: clean vs dirty segments, the cleaner thread and its offset map, tombstones and delete.retention.ms, dirty-ratio and…

Read article
ARTICLE · 45

Kafka MirrorMaker 2 architecture

Deep-dive on Kafka MirrorMaker 2 (MM2): cross-cluster replication built on Kafka Connect that copies records, topic configs, ACLs, and consumer offset…

Read article
ARTICLE · 46

Apache Paimon architecture

Deep-dive on Apache Paimon: LSM inside, streaming ingest with Flink, batch reads, merge engines, change-log stream, snapshots.

Read article
ARTICLE · 47

Apache Pulsar Architecture in Depth

A 2500-word walkthrough of Pulsar: brokers, BookKeeper bookies, ZooKeeper, topics, subscriptions, geo-replication, tiered storage, Functions.

Read article
ARTICLE · 48

Redis Streams Architecture in Depth

A 2500-word walkthrough of Redis Streams: XADD, entry ID, consumer groups, XREADGROUP, XACK, XPENDING, XCLAIM, MAXLEN, persistence + cluster.

Read article
ARTICLE · 49

Schema registry architecture

Deep-dive on the streaming schema registry: subjects and versions, the schema-id wire format, backward/forward/full compatibility modes, reader vs wri…

Read article
ARTICLE · 50

Stateful stream joins -- joining unbounded streams

Deep-dive on stateful stream joins: the unbounded-join problem, keeping both sides as state (buffered by key), window/interval joins bounding the stat…

Read article
ARTICLE · 51

Kafka tiered storage architecture

Deep-dive on Kafka tiered storage: the RemoteLogManager and remote index, segment rolling and upload, local vs remote retention, tail vs backfill read…

Read article
ARTICLE · 52

Stream windowing -- bounding the unbounded

Deep-dive on stream windowing: the unbounded-stream aggregation problem, tumbling/sliding/session windows, event vs processing time, watermarks for ev…

Read article