{"id":13577,"date":"2022-07-19T13:51:37","date_gmt":"2022-07-19T20:51:37","guid":{"rendered":"https:\/\/www.couchbase.com\/blog\/?p=13577"},"modified":"2023-05-22T06:21:54","modified_gmt":"2023-05-22T13:21:54","slug":"couchbase-capella-vs-cosmosdb-nosql","status":"publish","type":"post","link":"https:\/\/www.couchbase.com\/blog\/couchbase-capella-vs-cosmosdb-nosql\/","title":{"rendered":"Comparing Couchbase Capella vs CosmosDB"},"content":{"rendered":"

CosmosDB is Microsoft’s NoSQL offering that’s exclusive to Microsoft Azure. It used to be called DocumentDB, but they changed the name and added some interesting new features. Let\u2019s go a little deeper on it and explore its strategy, documentation, what developers have been talking about, and how it compares to Couchbase Capella.<\/span><\/p>\n

One Database to rule them all?<\/b><\/h3>\n

Microsoft claims that CosmosDB is a NoSQL database able to do literally <\/span>everything<\/span><\/i>: It is a Document database, Columnar storage, a Key-Value Store and a Graph Database. All achieved thanks to an abstraction of the data format called <\/span>atom-record-sequence (ARS)<\/span><\/a>.<\/span><\/p>\n

Let’s look at how data is organized according to each model. First, you have to choose the API you would like to use (SQL, MongoDB API, Microsoft Azure Table, Cassandra or Gremlin) and stick with it, as it can\u2019t be changed later. But behind the scenes, it looks to be a <\/span>custom JSON format<\/span><\/a>.<\/span><\/p>\n

CosmosDB is trying to compete with all of the major NoSQL databases, which may be a risky strategy. For one, this approach may limit the features that CosmosDB can ultimately offer. There’s a single common denominator, and that can’t be strayed too far from. Also, APIs like MongoDB and Cassandra are not defined or planned by Microsoft. This means that Microsoft will always be catching-up to the latest releases, and will ultimately never achieve 100% compatibility. Microsoft maintains documentation about <\/span>which MongoDB features are supported<\/span><\/a> and which are not (and the same thing for <\/span>Cassandra<\/span><\/a>). An <\/span>all-in-one<\/span><\/i> solution like CosmosDB might be good for simple applications with few functionality demands, but all those abstractions come with a cost and will ultimately impact simplicity, performance and be feature limited.<\/span>\u00a0<\/span><\/p>\n

Couchbase vs CosmosDB \u2013 Comparing Apples with \u201cApples\u201d<\/b><\/h3>\n

\u00a0<\/span>This comparison will focus most on scenarios that make sense to compare both technologies (for example, Couchbase is not a graph database, so the comparison wouldn’t make sense).<\/span><\/p>\n

One other important note: Couchbase Capella is Couchbase’s DBaaS (database-as-a-service) offering, available in AWS and GCP (soon to be in Azure too). It is basically a managed version of Couchbase Server, which is still available for download, so they are very similar. Unless otherwise stated, the “Couchbase” column applies to both Capella and Server.<\/span><\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Feature<\/span><\/td>\nCosmosDB<\/span><\/td>\nCouchbase Capella<\/span><\/td>\n<\/tr>\n
Licensing<\/span><\/td>\nProprietary, closed-source but free-tier is available.\u00a0<\/span><\/td>\nFree trial available for Capella<\/span><\/a>, <\/span>Couchbase Community and Enterprise available for download<\/span><\/a>, <\/span>BSL<\/span><\/a><\/td>\n<\/tr>\n
Type<\/span><\/td>\n\n
    \n
  • Key-value<\/span><\/li>\n
  • Document<\/span><\/li>\n
  • Graph<\/span><\/li>\n
  • Columnar<\/span><\/li>\n<\/ul>\n<\/td>\n
\n
    \n
  • Key-value<\/span><\/li>\n
  • Document<\/span><\/li>\n
  • Built-in cache<\/span><\/li>\n
  • Mobile<\/span><\/li>\n<\/ul>\n<\/td>\n<\/tr>\n
Model<\/span><\/td>\n\n
    \n
  • 2MB limit per document<\/span><\/li>\n
  • 16MB limit for Mongo mode only<\/span><\/li>\n<\/ul>\n<\/td>\n
\n
    \n
  • 20MB document limit<\/span><\/li>\n<\/ul>\n<\/td>\n<\/tr>\n
Search<\/span><\/td>\n\n
    \n
  • Requires a separate, propietary product: Azure Cognitive Search<\/span><\/li>\n<\/ul>\n<\/td>\n
\n
Indexing<\/span><\/td>\n\n\n
Data Integrity<\/span><\/td>\nFive options are available in configuration:<\/span><\/p>\n
    \n
  • Strong<\/span><\/li>\n
  • Bounded staleness<\/span><\/li>\n
  • Session (default)<\/span><\/li>\n
  • Consistent prefix<\/span><\/li>\n
  • Eventual<\/span><\/li>\n<\/ul>\n<\/td>\n
\n
    \n
  • Strong consistency<\/span><\/li>\n
  • Query consistency can be specified on a per-query basis<\/span><\/li>\n<\/ul>\n<\/td>\n<\/tr>\n
Scalability<\/span><\/td>\nHighly scalable<\/span><\/td>\nHighly scalable<\/span><\/a><\/td>\n<\/tr>\n
Mobile<\/span><\/td>\nNo plans for CosmosDB for mobile or devices<\/span><\/a> or any offline support<\/span><\/td>\n\n
    \n
  • Couchbase Lite provides a mobile\/device\/edge database.<\/span><\/li>\n
  • Sync Gateway automatically syncs to\/from the data center<\/span><\/li>\n<\/ul>\n<\/td>\n<\/tr>\n
Deployment<\/span><\/td>\nAzure only, fully managed only.<\/span><\/p>\n

There is a development version available (currently Windows only).<\/span><\/td>\n

Can be deployed anywhere, including Azure, on-premises, Kubernetes, Docker, VM, bare-metal.<\/span><\/p>\n

Couchbase Capella offers a fully managed DBaaS<\/span><\/td>\n<\/tr>\n

Locking<\/span><\/td>\nOptimistic and pessimistic locking available<\/span><\/td>\nOptimistic and pessimistic locking available<\/span><\/td>\n<\/tr>\n
Backup & Restore<\/span><\/td>\nContinuous backup mode for 30 days<\/span><\/p>\n

Periodic backup mode (default)<\/span><\/td>\n

Automatic <\/span>backup and restore<\/span><\/a> service with configurable backup wizard<\/span><\/p>\n

Continuous backup available using XDCR<\/span><\/td>\n<\/tr>\n

Querying<\/span><\/td>\nBased on which mode is chosen.<\/span><\/p>\n

Example 1: SQL API is an extremely limited subset of standard SQL<\/span><\/p>\n

Example 2: MongoDB API is a non-100% subset of Mongo API<\/span><\/td>\n

Full SQL implementation<\/span><\/a> called SQL++ (with JOIN, aggregate, CTE, window functions, CRUD operations, etc) – previously known as “N1QL”<\/span><\/td>\n<\/tr>\n
Data Center Replication<\/span><\/td>\nPush-button global master-master replication between supported Azure data centers<\/span><\/td>\nXDCR<\/span><\/a> allows any combination of unidirectional and bidirectional replication between any Couchbase deployment, including data filtering<\/span><\/td>\n<\/tr>\n
Speed\/performance<\/span><\/td>\nMore speed and performance is only obtained by increasing <\/span>RUs<\/span><\/a>, which will often be prohibitively expensive\u00a0<\/span><\/td>\nMemory-first read and write operations.<\/span><\/p>\n

Built-in caching layer.<\/span><\/p>\n

Can be tuned by increasing memory, disk, or adding a new node.<\/span><\/p>\n

Memory-optimized indexes available<\/span><\/td>\n<\/tr>\n

Sharding \/ partitioning<\/span><\/td>\nPartition key(s) must be created and managed manually, requiring a dedicated expert to set and design correctly in order to reach performance\/scale goals<\/span><\/td>\nSharding is <\/span>completely automatic<\/span><\/a><\/td>\n<\/tr>\n
Architecture<\/span><\/td>\nUnknown \/ proprietary<\/span><\/td>\nEvery node is a master<\/span><\/a> in Couchbase, making most efficient use of resources<\/span><\/td>\n<\/tr>\n
Supported SDKs<\/span><\/td>\n.NET (primary, most feature complete)<\/span><\/p>\n

Other SDKs:<\/span><\/p>\n

    \n
  • Java<\/span><\/li>\n
  • Node.js<\/span><\/li>\n
  • Python<\/span><\/li>\n<\/ul>\n

    \u00a0\u00a0\u00a0\u00a0\u00a0(Others through Mongo\/Cassandra)<\/span><\/td>\n

.NET<\/span>C \/ C++<\/span><\/p>\n

Go<\/span><\/p>\n

Java<\/span><\/p>\n

Node.js<\/span><\/p>\n

PHP<\/span><\/p>\n

Python<\/span><\/p>\n

Ruby<\/span><\/p>\n

Scala<\/span><\/p>\n

Kotlin<\/span><\/td>\n<\/tr>\n

<\/td>\n<\/td>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Success in the Real World<\/b><\/h3>\n

This side-by-side comparison may favor Couchbase, but what about the real-world experiences of an organization that was using CosmosDB and switched to using Couchbase?<\/span><\/p>\n

Facet Digital cut their database costs<\/span><\/a> by 50%, and improved their performance by 100x by switching to Couchbase Capella.<\/span><\/p>\n

How was it possible?<\/span><\/p>\n