The open source database maker Yugabyte has released a public database-as-a-service version of its YugabyteDB product called Yugabyte Cloud. YugabyteDB is a distributed Postgres-compatible SQL database that provides multiple APIs for different developer needs.
“Our goal is to simplify the database tier for building cloud-native transactional applications,” said Karthik Ranganathan, co-founder and CTO of Yugabyte.
The prices for Yugabyte Cloud are based on the number of vCPUs used in the backend system, with usage-based prices of 0.25 USD / vCPU / hour with a billing increment of 1 minute. The annual subscription price offers a discount on the PAYG prices.
Yugabyte also has a “perpetual free tier” for developers to try things out before committing to the platform. Customers are limited to a single-node cluster that is not suitable for production workloads, but Yugabyte says it has enough resources to begin exploring its core functionality.
It will be interesting to see how the offering compares to Amazon RDS for PostgreSQL and Azure Database for PostgreSQL – Hyperscale (based on CitusData, acquired by Microsoft in 2019).
Developers tend to manipulate their data through an API that suits the nature of their workload. Working with network data tends to prefer some kind of graph database, while massive or more object and document style data tend to prefer a NoSQL approach. Still others do well with the more traditional relational database and SQL query approach.
“We want to make sure that we break the barriers to availability and scalability no matter which API you choose, while delivering the value that the API itself offers,” said Ranganathan.
Yugabyte’s goal is to essentially remove the need for developers to worry about the database layer and simply manipulate data using the APIs they think are best for what they want to do. It’s another example of a company that specifically targets developers as the primary audience for its software, rather than the infrastructure teams that traditionally control what database systems are available to developers.
A cloud option makes a lot of sense for Yugabyte to promote the acceptance of its database. People who just want to try it out don’t necessarily want to spend a lot of time setting up their own lab environment when they can quickly build a free internet-based database to play around with. It also bypasses a lot of the internal friction and gatekeeping that can keep people from trying new things. It’s the same playbook that made AWS successful in the first place.
A major challenge for Yugabyte is the sheer number of database options that exist, many of them open source and with well-developed ecosystems around them. Choosing a database is not just about a place to store your data, it is also about the tools needed to run the system, optimize performance, and so on. Being PostgreSQL compatible should help, but it depends a lot on how tight the compatibility is.
When we spoke, Ranganathan himself was cautionary story by RethinkDB and that being technically better is not enough to see customer acceptance. Ease of use for solving actual problems is more important.
“We actually have the opposite problem of getting people to understand the nuances of the difference between Postgres and a distributed SQL database and the tradeoffs,” he said.
Rather than targeting the new builders, the hobbyists, and the less important systems, Yugabyte is targeting the systems that have been built to be mission-critical, but in both cases are reaching inflection points in scale, performance, operations management, or other cloud options on-premise. These are people who are already ready to switch to something new because they are currently in pain and are looking for relief.
Yugabyte wants to relieve this pain and become a viable option for others to avoid this pain in the first place.