Azure Documentdb vs MongoDB

(This article is also published in my public number "dotnet daily Essence article")

The articles recommended today compare Azure Documentdb and MongoDB, as well as some suggestions for use.

I think a lot of friends should know about Azure, but there are also a lot of people who don't know that Microsoft has its own NoSQL database product. Of course, this product is only available in PAAs for use in azure. Documentdb started previewing last fall, this is officially on-line (according to Azure, generally Available). Of course, in the NoSQL market, melee is fierce, especially MongoDB accounts for nearly half of the market share. David Green, the author of this article, specifically compared Documentdb with MongoDB.

Same point

He first analyzed the similarities between the two. NoSQL databases are typically divided into the following categories: Key-value storage, Columnstore, graph storage, document storage, or hybrid models. Both Documentdb and MongoDB share the same categories of document storage, with the following characteristics: using partitioning to fault tolerance, inverse paradigm to organize data sets to avoid the traditional relational structure, imitation object-oriented programming in the entity object to highlight human-readable data format.

In addition to these common features above, both have some special advantages. First, both support client Sdk,documentdb in multiple programming languages support mainstream. NET, node. js, JavaScript, Python, and Java, and of course mangodb support is richer. Second, both use a similar data format--json (MongoDB uses the extended Bson of JSON). Finally, the common language used by both is JavaScript.

Differences and advantages

After talking about the similarities between the two, David highlights some of the advantages of the differences and DOCUMENTDB.

• Paas:documentdb is provided directly from PAAs. This brings the benefits of simpler configuration, management, and maintenance. MongoDB needs to be deployed to VMS on its own, and costs are spent on operations. Because of the many benefits of PAAs, the authors recommend using a third-party ready-made PAAs even if you want to use MongoDB.
• Scaling capability: Because DOCUMENTDB is a PAAs-driven architecture, it handles horizontal scaling in a completely different way from MongoDB. There is no need to manage replication after the Documentdb partition, and MongoDB needs to process replication at the same time. This is also the benefit of documentdb back-end dependent on the scaling capabilities of azure.
• Native rest interface: While both provide developers with multiple language SDKs, the Documentdb is native to the rest interface, but the SDK is also the wrapper for the rest interface. In contrast, MongoDB does not have a native rest interface, but it has wire protocol and metadata-driven (TCP-based) access to data that can be language-independent. In some cases, however, HTTP-based rest interfaces are obviously more convenient (such as the Internet of Things).
• Data Interchange Format: DOCUMENTDB uses JSON more standard (RFC 7159 and ECMA-404).
• Index processing: Both are indexed based on B-tree, but DOCUMENTDB provides two types of index hashes and range,range that temporarily do not support time fields, DOCUMENTDB also does not support indexing of geo-location information but relies on azure Search to solve this problem. From the product point of view, MongoDB has the advantage, but the actual use of the process will not be too big problem.
• Asynchronous processing: Because DOCUMENTDB provides a rest interface, these interfaces, or the. NET SDK, provide async/await support to provide concurrency processing power.
• Pricing: Although MongoDB is open source for free, the cost of operation is not small. Documentdb is paid based on usage, but is not expensive and can be obtained through DreamSpark and BizSpark for azure free subscriptions.
• Consistency: MongoDB consistency can be configured to enable consistency, and DOCUMENTDB can configure a level 4 consistency level.
• Binary Large object storage: MongoDB relies on Gridfs to implement BLOB storage, DOCUMENTDB relies on azure blob Storage.
• Monitoring: Azure provides rich monitoring metrics for DOCUMENTDB, and MongoDB tracks host hosts through MONGO monitoring Service (MMS).
• Programmability: Both support the JAVASCRIPT,DOCUMENTDB. NET SDK for better LINQ support, but not for debug support (mainly without a local emulator).
• Other differences: Documentdb There is a temporary limit to the aggregation operation, no service-side ordering, the tool is not rich enough. MongoDB situation is slightly better.

Finally, the author gives some suggestions for use: Documentdb can meet the basic needs of nosql, but documentation and application cases need to continue to improve. Anyway, documentdb can get you (in the case of Azure) quickly into the world of NoSQL.

Because of the huge amount of information, more details are "read the original".

Original address: http://justazure.com/mongodb-vs-azure-documentdb/

Azure Documentdb vs MongoDB