Snapshot isolation

[Andrea_Scotti]

Hi.
When operating a read with snapshot isolation, the transaction is guaranteed to get the latest one?
Thanks

[ptpaterson]

Hi @Andrea_Scotti.

In the case of a non-serialized index (with snapshot isolation), the transaction will get the latest data, at the time of request. Any reads to the index within the same transaction will be consistent. However, it just won’t lock the data from being changed by other incoming requests.

Isolation levels are outlined in the docs here:

https://docs.fauna.com/fauna/current/concepts/isolation_levels.html

There is a detailed discussion about it in practice in the forums – not sure if it will help your immediate question about snapshots, though.

Concurrency and isolation levels

I am trying to further understand Fauna’s isolation levels and concurrency, but I am not sure I am thinking about it correctly. Say I have an application that supports 2 concurrent users, both reading from and writing to the same document in a collection. This simple starting document is enough to illustrate the example: { selections: [] } Now let’s look some FQL that is responsible for handling user actions. In summary, this FQL does the following: Reads the document Performs some compl…

[Andrea_Scotti]

Ok then, I understand.
I knew that, theoretically, snapshot isolated reads do not guarantee you get the latest snapshot of a piece of data. You get a *certain snapshot…

Reading dgraph docs, which I know does not guarantee any isolation above snapshot, nonetheless I found a nice explanation of guarantees about getting last available version of any data under snapshot isolation, and thus wondered how Fauna looks like in this respect.

https://dgraph.io/docs/design-concepts/consistency-model/

Any comment is appreciated,
Thanks