Irrespective of the different noSQL databases that have appeared (Key-value, documentaries, columnars, networks, time series and content repositories), a new concept has emerged in the last few years: the databases in the cloud. These projects have been developed by large companies such as Microsoft or Google, and their main feature is that they offer data storage as a service.
These supports are gaining market share mainly for one reason: NoSQL clustered servers are complex to maintain and generally expensive, both in terms of infrastructure and skilled personnel. These data services offer maintenance and support included in the price.
But their advantages do not end there. All of these platforms have more fully integrated associated services (data processingThe most common models are the "cloud" (e.g., connectivity with multiple platforms through APIs, etc.), and security systems maintained and endorsed by the companies themselves. In addition, they tend to offer mixed models. In other words, they are convenient, easy to maintain and facilitate the creation of multi-application environments. On the downside, these platforms are fee-based, and often this fee is calculated on the basis of both storage and read/write usage, which can increase the price of maintenance. All of them offer free versions, but these are limited to low volumes of use.
Main NoSQL databases in the cloud
Amazon DynamoDB (AP)
Managed by Amazon Web Services (AWS), supporting key-value and document models. Contains Amazon DynamoDB Accelerator (DAX), an in-memory cache to reduce response times. Sharding partitioning. MapReduce can be implemented using Amazon EMR, but does not come standard. Access controls defined using AWS identity and Access Management (IAM).
Microsoft Azure Cosmos DB (CP)
Integrated in the Microsoft Azure system. It allows the implementation of document, key-value, columnar and even graph models. It has APIs for different physical SQL and NoSQL databases, as well as connectors with other Azure applications such as Databricks Spark, GemlinAPI (graphs) or Azure Functions. Access rights can be defined at item level.
Google Cloud DataStore (CP):
Managed by Google. It offers a scalable document data model that can be easily integrated with multiple Google applications, such as App Engine or Datastore. It has the possibility of ACID transactions and SQL queries with Google Query Language, as well as implementing access controls based on Google Cloud Identity and Acess Management (IAM).
