Topics to Learn
Toggle theme
Message Broker
Kafka
used as key value but stored as binary in kafka
default port
serialisation and deserialization
pros and cons
Kafka cluster
Fault Tolerance
Scalability
Distributed Processing
Kafka Broker
topics
compacted topics
partitions
leader
follower
replication
replication factor
key
segments
Producer
record
header
key
value
timestamp
retention period
ack /nack
no acks
leader acks
all acks
Consumer
Queue vs Pub Sub
Consumer group
Offset
Connectors
At most once
At least once
Exactly once
Exactly-Once Semantics
Idempotent
Two-Phase Commit
alt
Persistent storage
Steam processing
Distributed system
leader
follower
zoo keeper
Metadata Management
Leader Election
Synchronisation
Heartbeats and Timeouts
Monitoring
default port
gossip
long polling
Kafka Connect
RabbitMQ
TCP
HTTPv2
AMQP
RabbitMQ server
default port
Exchange Queues
Heartbeats
Connection pool
Channels
Multiplexing
Concurrency
Message TTL
Message Acknowledgment
Strategies
Automatic Acknowledgment (Ack)
Positive Acknowledgment
Negative Acknowledgment (Nack)
Rejection with Requeue
Rejection without Requeue
Exchanges
Fanout exchange
pros and cons
uses
Direct exchange
pros and cons
uses
Header exchange
pros and cons
uses
Topics exchange
pros and cons
uses
Dead Letter Exchanges and Queues
Polyglot persistence
Durability
Durable Queues
Persistence message
Combined Durability
rabbitMQ
Routing Key
Request response
architecture
breaks
pros and cons
Publish subscribe (pub/sub) model
Queue/Channels/Topics
Publisher/producer
Consumer
pros and cons
Multiplexing
Channel
Push model
gRPC
why?
http
protobuffer
Unary gRPC
Server streaming
Client streaming
Bidirectional