It removes the most painful part (in my opinion) of the relational database, which is the operation. As a developer looking to just develop, it gives me the abstractions and tools I need to develop the data store my app needs, without having to worry so much about how I will operate, patch, and scale the database. This isn't to say that database knowlege is not required at all, since it will happily let you write horribly inefficient queries like most databases.

The recent introduction of schema enforcement to collections (tables in RDBMS parlance) along with custom check constraints makes it that much easier to migrate from your existing RDBMS. The flexibility of having schemaless or partially typed collections means that prototyping is easy and allows clamping down on the schema over time as the requirements become solidified.

The ability to run logic operations, assign variables, do math, perform list/set transformations, and other things you would normally do in code inside the database makes it particularly well suited to a serverless environment. You can keep all of the transactional business logic within Fauna so that it is truly transactional, with much less application code to handle what happens if a transaction fails part way through. It is also regionless, meaning any edge compute service you may use will automatically connect to the closest replica to keep request times down for a globally distributed user base. Note that they do offer USA, EU, and Global database options with slightly different usage based pricing.

Their free tier is also generous enough to prototype and run apps in order to determine if it's worth scaling the usage throughout your organization. Review collected by and hosted on G2.com.

Unlike a traditional SQL RDBMS, you cannot span a single transaction over multiple requests. This means that all of the compute operations that might happen in between different SQL calls inside of a transaction have to be translated into FQL expressions and run inside the database. This means it takes a more work to port an existing app that uses a SQL RDBMS to Fauna.

Though Fauna is serverless in operation, the costs do not "scale to $0" like some other serverless offerings. Once you exceed the limits within their generous free tier, you have to jump into a pay as you go plan or one of the other plan tiers. These have a flat monthly rate + overage model, so you pay every month regardless of how much you use. Being forced to jump into a higher tier because you want some feature locked to that higher tier while your actual usage doesn't come close to the monthly fee doesn't feel great. I would have preferred it to be a truly pay for what you use model. Review collected by and hosted on G2.com.

When I say "best of every world", it's hard to quantify. But when I talk to people about Fauna and explain it's a globally distributed, fully serverless, ACID compliant, document based relational database they always say something alongs those lines.

Fauna has solved so many problems for us. It has the flexibility of a document based database and the power, security and reliability of relational database. We've been able to streamline our code, reduce response times and add powerful new features as a result of replacing DynamoDB with Fauna.

Not only that, but its a breeze to set up and get the ball rolling. The FQL query language is simple, but powerful. If you can write code, you can write FQL. There's no learning curve to understand a new query language. There are some quirks to gets to grips with, but we've worked closely with the team at Fauna and they've been amazing.

We believe that Fauna is the future of databases, and that when you decide to give it a try you'll agree with us. Review collected by and hosted on G2.com.

Fauna is a comparitively young DB technology, so don't expect a bustling commiunity answering questions on StackOverflow like you might with an RDBMS. But the documentation is clear and concise, and updated regularly. Review collected by and hosted on G2.com.

The combination of being able to securely run queries on the client or an edge device with Javascript-like queries to get and mutate data make Fauna unique in the database space. Developers can quickly pick up the new tool and easily modify access rules centrally. The serverless nature of the service is amazing and is the main reason you can easily connect a massive set of devices (or lambdas) without worrying about connection pooling. Review collected by and hosted on G2.com.

There is nothing to dislike about Fauna but because the product is so new, it is hard to find well-documented patterns for it. Reaching out to the Fauna team is the best way to resolve edge cases. Review collected by and hosted on G2.com.

We don't have to manage any infrastructure, the cost scales with us

The FQL X syntax is very similar to TypeScript, which means we don't have to develop across different paradigms (frontend, backend, and database all use the same syntax)

Schema management built-in means we don't have to worry about generating/applying/rolling back schema migrations

Close partnership, and active Discord, means any questions or challenges are addressed quickly and time is taken by technical people to address them Review collected by and hosted on G2.com.

Although you can setup private endpoints to connect privately via AWS, I can't find any documentation on doing the same for on-premise architecture. Fauna is doing a great job at integrating with new cloud technologies, but it might not be an easy option for some types of infrastructure deployments. Review collected by and hosted on G2.com.

It's managed, serverless, and highly scalable. The ability to write typescript like queries, and have all the reads and updates be globally consistant is amazing. The ease of projecting extra data greatly reduces the need for round-trip queries which makes everything that much more performant. Finally, the fine-grained security makes it easy to allow the client to directly query the database! Review collected by and hosted on G2.com.

The type system can be difficult to debug. Also, fauna lacks full text search. Review collected by and hosted on G2.com.

Highly reliable, scalable and low latency serverless database. Its document-based renational schema makes it highly flexible for different use cases. Review collected by and hosted on G2.com.

The learning curve and the lack of more advanced data exploration features can add friction. Review collected by and hosted on G2.com.

Fauna has a simple to use user authentication and authorization system that is at the same extendible to a powerful enterprise ABAC system.

Also, I don't need to worry about managing the server because it's serverless.

Then you need to know that it's a document-based database. You store documents in a JSON structure, which gives you a lot of flexibility over traditional SQL table databases. In contrast to other document databases, Fauna offers Schemes and Types, so you always have a clear picture of your data, including data integrity. Review collected by and hosted on G2.com.

They could explain their great product better and more simply. You need some time to get into the docs, but you will be highly rewarded if you do it. Review collected by and hosted on G2.com.

Fauna sure does offer a lot of functionality in their free version. They have plenty of SDKs to choose from and a lot of walkthrus and tutorials. The promise of trasanctional nosql with indexes is a strong. Review collected by and hosted on G2.com.

The learning curve on Fauna looks daunting. It has the feel of MongoDB at first. However, it might be too much if you come from a pure SQL background. I get the feeling that Fauna might be the last database you will ever need, if you can figure out how to use it. Review collected by and hosted on G2.com.

Fauna DB is fast in spinning up databases and connecting to them, though it is a serverless DB solution.

There are rarely any cold starts when using it.

The FQL language they have developed is very similar to GraphQL which is easy to learn for someone new to Fauna DB. Review collected by and hosted on G2.com.

Even though FQL is easy to learn and native support for GraphQL is there, a person coming from SQL background will find it complicated and intimidating to use.

If native support for this is added, it will be really helpful to SQL devs. Review collected by and hosted on G2.com.

Robust, fast, no downtimes, good balance between no-schema documents and indices.

You can manage users and roles in the db, and alegedly call it directly from your client applications. I personally don't use this because my users are managed in firebase, and I have an API that facilitates all calls to the DB Review collected by and hosted on G2.com.

Creating indices for everything, steep learning curve for FQL and user-defined functuons.

A lot of the app code goes into the DB as user-functions. I find it hard to maintain my app's code this way. I would prefer to have all my code in the server that calls the db. Review collected by and hosted on G2.com.