Jonathan Ellis

A common misconception is that masterless databases like Cassandra are designed to tolerate network partitions, which are&nbsp;<a href="https://twitter.com/kellabyte/status/391076376955998208">quite possible</a>&nbsp;but relatively uncomon within a single data center compared to other failures.

A more complete understanding is that Cassandra is designed to tolerate&nbsp;any&nbsp;failure without interruption, whether due to node failures, garbage collection pauses, or slow disks,&nbsp;<a href="http://pl.atyp.us/wordpress/index.php/2009/11/availability-and-partition-tolerance/">all of which are indistinguishable from partitions to the rest of the system</a>.
<img alt="Cassandra Fault Tolerance" data-align="center" data-entity-type="file" data-entity-uuid="0d062e59-7b49-40a1-ad33-ec38a85784c9" src="https://www.datastax.com/sites/default/files/inline-images/rrp-700x399.png" />
Cassandra fault tolerance. Click for details

An even more complete understanding is that designing to accomodate failure also drives latency lower: since there are no masters and any replica can answer requests, Cassandra has no artificial bottleneck that makes latency spike whenever there is a hiccup.

HBase may be the most well-known master-oriented distributed database competing with Cassandra. While HBase developer Enis Soztutar reports that "<a href="https://issues.apache.org/jira/browse/HBASE-10070">[i]n the present HBase architecture, it is hard, probably impossible, to satisfy constraints like 99th percentile of the reads will be served under 10 ms,</a>"&nbsp;Cassandra easily achieves this in many clusters.

Here's one example, from a production cluster in England (in microseconds):
<img alt="Production Cluster Example" data-align="center" data-entity-type="file" data-entity-uuid="1bbaf377-5940-4f6f-ad78-b2edf60ff4ea" src="https://www.datastax.com/sites/default/files/inline-images/99th.png" />
Part of the explanation is that Cassandra is simply&nbsp;<a href="http://vldb.org/pvldb/vol5/p1724_tilmannrabl_vldb2012.pdf">faster overall</a>. But more broadly, Cassandra's fault tolerant design is what allows the slowest 1% of requests to be less than twice as slow as the average.

Cassandra vs failures, latency edition

Jonathan EllisTechnology

Share

More Technology

Getting Started with DataStax Astra DB and Amazon Bedrock

Why You Can't Miss Cassandra Summit

Mitigate the OpenAI Debacle with Open Source and Cloud

Supercharge AI Assistants with Superagent and DataStax

NoSQL and Vector DB
for Generative AI,
Instantly, At Scale