dkduckkit.dev

Platform Engineering Glossary

Definitions for Kafka, rate limiting, latency, and API versioning terms used by platform engineers. Click any term to learn more with detailed explanations, formulas, and practical examples.

Latency & SRE

Latency budget

Total time allocated for a complete user-facing request across all architectural hops.

Try in 1 tool
Latency & SRE

SLA and SLO (service level agreement vs objective)

SLA is a contract with guarantees; SLO is the internal target set stricter than SLA to create error budget.

Try in 1 tool
Latency & SRE

p50 latency (median)

50th percentile of request durations: half of requests complete faster, half slower.

Try in 1 tool
Latency & SRE

p99 latency

99th percentile of request durations: captures experience of users under stress.

Try in 1 tool
Latency & SRE

Headroom (latency)

Gap between measured latency and SLA ceiling; buffer for unpredictable spikes.

Try in 1 tool
Latency & SRE

Cold start (serverless)

Extra latency when serverless function starts on new execution environment.

Try in 1 tool
Latency & SRE

Cross-AZ latency

Round-trip time between availability zones; adds cost and latency to replication.

Try in 1 tool
Latency & SRE

Tail latency

High-percentile latency values (p99, p99.9) representing slowest requests.

Try in 1 tool
Latency & SRE

Fan-out (distributed systems)

Single request triggers multiple parallel downstream calls.

Try in 1 tool
Latency & SRE

N+1 query problem

One query to fetch list plus N queries for each item; causes database round-trips.

Try in 1 tool
Latency & SRE

Speed of light in optical fibre

~200,000 km/s; physical lower bound on network latency regardless of optimization.

Try in 1 tool
Latency & SRE

Thundering herd

Many clients simultaneously access resource that just became available.

Try in 1 tool
Latency & SRE

Percentile latency (p50 / p99 / p999)

Statistical measure of request duration distribution.

Try in 1 tool
Rate Limiting

Fixed window rate limiting

Counts requests in discrete, non-overlapping time buckets.

Try in 1 tool
Rate Limiting

Sliding window rate limiting

Maintains rolling count of requests over recent N seconds.

Try in 1 tool
Rate Limiting

Token bucket

Accumulates tokens at fixed rate; requests consume tokens.

Try in 1 tool
Rate Limiting

Leaky bucket

Processes requests at constant output rate, queueing excess.

Try in 1 tool
Rate Limiting

Burst limit

Maximum requests above steady-state rate in short period.

Try in 1 tool
Rate Limiting

Rate limit window

Time period over which request counts are measured.

Try in 1 tool
Rate Limiting

Global vs per-consumer rate limit

Global limit shared across all clients vs per-consumer independent counters.

Try in 1 tool
Rate Limiting

Jitter (retries)

Adds randomness to retry wait times to prevent synchronized retries.

Try in 1 tool
Rate Limiting

Exponential backoff

Retry wait times grow multiplicatively after each failure.

Try in 1 tool
Rate Limiting

Retry storm

Failure amplification where synchronized retries recreate overload.

Try in 1 tool
Rate Limiting

HTTP 429 Too Many Requests

Standard status code for rate limiting responses.

Try in 1 tool
Rate Limiting

RateLimit headers (IETF)

Standardized headers for communicating quota information.

Try in 1 tool
Rate Limiting

X-RateLimit headers

Legacy de facto standard for rate limit state communication.

Try in 1 tool
Rate Limiting

Retry-After header

Specifies how long client should wait before retrying.

Try in 1 tool
Rate Limiting

API throttling vs rate limiting

Rate limiting rejects excess; throttling delays to preserve availability.

Try in 1 tool
Rate Limiting

nginx limit_req

Nginx directive implementing leaky bucket rate limiting (token-bucket-like with nodelay).

Try in 1 tool
Rate Limiting

Kong rate-limiting plugin

Enforces request quotas on routes and services.

Try in 1 tool
Kafka

Kafka partition

Fundamental unit of parallelism and ordering within a topic.

Try in 2 tools
Kafka

Consumer lag (Kafka)

Difference between latest message and consumer committed offset.

Try in 2 tools
Kafka

Consumer group (Kafka)

Set of consumers cooperatively consuming a topic.

Try in 2 tools
Kafka

Consumer offset (Kafka)

Position consumer group has committed as processed.

Try in 2 tools
Kafka

Effective consumers (Kafka lag)

Actual consumers processing partitions, capped at partition count.

Try in 2 tools
Kafka

Consumer poll loop (Kafka)

Central control loop: fetch records, process, repeat.

Try in 2 tools
Kafka

Consumer rebalance (Kafka)

Redistributing partition assignments among group consumers.

Try in 2 tools
Kafka

Replication factor (Kafka)

Number of brokers storing copies of each partition.

Try in 2 tools
Kafka

Log retention (Kafka)

How long messages kept before deletion eligibility.

Try in 2 tools
Kafka

Producer compression (Kafka)

Reduces batch size before sending to broker.

Try in 2 tools
Kafka

Throughput vs bandwidth (Kafka)

Actual processed rate vs theoretical maximum capacity.

Try in 2 tools
Kafka

max.poll.interval.ms

Maximum time between poll() calls before consumer considered dead.

Try in 2 tools
Kafka

max.poll.records

Maximum records returned by single poll() call.

Try in 2 tools
Kafka

Kafka batch.size

Maximum size of message batch before sending.

Try in 2 tools
Kafka

kafka linger.ms

How long producer waits to accumulate messages into batch.

Try in 2 tools
Kafka

replica.fetch.max.bytes

Maximum bytes follower fetches from leader per request.

Try in 2 tools
Kafka

message.max.bytes

Maximum size of single message batch broker accepts.

Try in 2 tools
Kafka

Kafka rebalance protocol (Eager vs Cooperative)

Eager stops all consumers; Cooperative only moves needed partitions.

Try in 2 tools
Kafka

CooperativeStickyAssignor

Incremental cooperative rebalancing with minimal partition movement.

Try in 2 tools
Kafka

Kafka partition skew

Uneven message distribution across partitions.

Try in 2 tools
Kafka

Kafka time lag

Elapsed time between message production and consumption.

Try in 2 tools
Kafka

Kafka claim check pattern

Stores large payloads externally, puts reference in Kafka.

Try in 2 tools
Kafka

Kafka RecordBatch

Wire format container grouping records for transmission.

Try in 2 tools
API Versioning

Semantic Versioning (SemVer)

MAJOR.MINOR.PATCH scheme communicating compatibility changes.

Try in 1 tool
API Versioning

Major, minor, and patch bumps (API)

Patch fixes bugs, Minor adds features, Major breaks compatibility.

Try in 1 tool
API Versioning

Breaking change (API)

Modification causing existing clients to fail or require updates.

Try in 1 tool
API Versioning

Calendar Versioning (CalVer)

Version tied to release dates rather than compatibility signals.

Try in 1 tool
API Versioning

URL versioning vs header versioning

URL embeds version in path; header uses custom headers.

Try in 1 tool
API Versioning

Sunset header (API)

Communicates when resource or version will be decommissioned.

Try in 1 tool
API Versioning

API deprecation strategy

Plan for retiring API while giving consumers migration time.

Try in 1 tool
API Versioning

Consumer-driven contracts

Consumers define expectations; provider runs contracts as tests.

Try in 1 tool
Databases

PgBouncer

Lightweight PostgreSQL connection pooler that multiplexes many clients onto a small pool of DB connections.

Try in 1 tool