Azure Table Storage vs CosmosDB Table API

Question

In Microsoft Build 2017 event, I came across CosmosDB Table API. It looks like Azure Table Storage.

Does it mean Microsoft is going to stop supporting Azure Table Storage eventually?

Now, Document Db is no more. It has been converted to CosmosDB now.

Also, the major problem with Table Storage is backup. Which has been resolved in CosmosDB Table API.

Also, pricing strategy for Table API is Throughput-based whereas for Table Storage is Storage-based.

So is Table Storage still a good bet for high volume applications?

Solution

As Peter responded, DocumentDB now is part of Cosmos DB, it is one of the many Data Models the new service has available including MongoDB API, Premium Tables API, and Gremlin Graph API.

They all have the same global distribution capabilities and consistency levels.

Now, going to your Tables question, both services have a different scenario and pricing model.

While Azure Storage Tables is aimed at high capacity on a single region (optional secondary read only region but no failover), indexing by PK/RK and storage-optimized pricing; Azure Cosmos DB Tables aims for high throughput (single-digit millisecond latency), global distribution (multiple failover), SLA-backed predictive performance with automatic indexing of each attribute/property and a pricing model focused on throughput.

Depending on your scenario is the service you might want to use, Azure Storage Tables is still a great solution, but if, for example, you need global distribution, Cosmos DB Tables is a best suited service.

Comments

[AlexB]

Good point on Azure Table Storage - not everyone needs to go global.

[Dogu Arslan]

not only global vs local but don't also forget about indexing on all properties. It is a pretty powerful feature and big leap in ways you can query your entity properties without being penalized for performance and scalability.

[Matias Quaranta]

Yeah, it was just an example. The previous paragraph enumerates several other features. Performance is another, Cosmos DB has a predictive high performance model with much better read/write times than Azure Storage Tables.

[Matias Quaranta]

The comment was probably removed due to its nature and moderators do that. The original question was if Table Storage was going to be removed and if it was good for high volume applications, see the Code of Conduct