Consistency vs. availability vs. latency
Where you meet it: quorums, replication modes.
What it is
Every mechanism in this article is also a placement on the trade-off between consistency (do readers see the latest write?), availability (does the system answer?) and latency (how fast?). Each mechanism moves the cost on that surface. None resolves it.
From Fault Tolerance & Delivery Semantics › Redundancy has no discount
The most important boundary in consistency is this one: causal consistency is the strongest model that is achievable with total availability. Above it you must coordinate. At or below it you need not. The unification and its price: CAP, PACELC, HAT, invariant confluence, CALM and external consistency starts from this line.
From Consistency & Isolation Models › Causal consistency
The classification matters less than the reframe: most systems' entire consistency cost is paid in the E branch, and CAP pointed everyone at the branch that almost never executes. To know what a consistency choice costs you, look at your p99 latency on a normal Tuesday, not at your partition runbook.
From Consistency & Isolation Models › CAP and PACELC
Where it appears
In 9 mechanisms. Each says what the mechanism is about, then how this trade-off shows up there, in its own words.
- Physical Substrate & Performance Laws
Why do trade-offs exist at all?
Synchronous replication across AZs costs ~1–2 ms per round trip.
Read it inTCP, kernel bypass and the latency tiers
- Transactions & Concurrency Control
What happens when many users change the same data at the same time, and how do we keep the result correct?
It converts atomicity from a correctness mechanism into an availability liability.
- Consistency & Isolation Models
When many users, many objects and many copies exist at once, which histories does the system promise never to show, and who writes the code for the histories it does show?
This is the most important boundary in this article: causal consistency is the strongest model that is achievable with total availability.
Read it inThe governing arithmetic: three resultsCausal consistencyThere is no true state any moreThe latency floor: one conservation law, four productsTunable consistencyExternal consistency, and what is verifiableCAP and PACELC, stated precisely
- Caching
What happens when we deliberately keep a second copy of data that no protocol forces to be correct?
Caching is concurrency control with the correctness requirement removed.
Read it inEvery entry is a claim about the pastThe first copy with no obligation
- Distributed Systems Coordination
Who is allowed to decide, and how does everyone else learn the decision without asking?
Anything whose staleness can only cost a retry belongs in discovery.
Read it inConsensusMembership, reconfiguration and discoveryNon-crash faultsReplication topologies: where is the serialization point?The arithmetic worth carrying
- Fault Tolerance & Delivery Semantics
What is the cheapest thing that can be safely done twice?
Both rules constrain the same three numbers, so choosing a quorum is choosing a point on the trade-off directly, with no slack.
Read it inReplication factor and quorums as a survivability budget
- Emergent Failure Dynamics
What happens when nothing is broken, every component is healthy, and the system's own response to stress makes the stress worse?
correctness > availability > completeness > freshness > latency
Read it inGraceful degradation and brownout
- Change Data Capture & Incremental Processing
What does it cost to know what changed, and is that cheaper than not needing to know?
This is idempotence purchased by restricting the consumer's semantics.
- Metadata, Catalog & Lineage
What can a system prove about its data without reading the data, and how does it know that proof is still true?
Metadata is a cache of facts about a system, kept in a different system, with no invalidation protocol.
Related
Made on the same pages most often.