Want smarter insights in your inbox? Sign up for our weekly newsletters to get only what matters to enterprise AI, data, and security leaders. Subscribe Now
Open source database company MariaDB plc today announced an update for SkySQL, its fully-managed cloud database service. Generally available on AWS and Google Cloud, the release comes a few months after the company’s public listing and brings two innovations to help teams better manage their cloud spend: autoscaling and serverless analytics.
MariaDB was created in 2009 as a software fork of Oracle-backed MySQL. The SQL-based relational database management system (RDBMS) started by offering high compatibility with MySQL, functioning as a drop-in replacement for the database in many cases. Currently, the open-source database (MariaDB Server) is maintained and offered by MariaDB plc, along with a commercial cloud database service called SkySQL.
SkySQL is fully managed, and enables teams to deploy and manage key MariaDB products — MariaDB Enterprise Server, the Xpand distributed SQL database, ColumnStore — in the cloud with just a few clicks.
Autoscaling and serverless analytics for MariaDB SkySQL
With the latest release, MariaDB says SkySQL gains the ability to autoscale both compute and storage in response to changes in demand.
AI Scaling Hits Its Limits
Power caps, rising token costs, and inference delays are reshaping enterprise AI. Join our exclusive salon to discover how top teams are:
- Turning energy into a strategic advantage
- Architecting efficient inference for real throughput gains
- Unlocking competitive ROI with sustainable AI systems
Secure your spot to stay ahead: https://bit.ly/4mwGngO
For instance, when CPU utilization tops 75% for all replicas for 30 minutes, a new replica or node will automatically be added to handle the increase. Similarly, when CPU utilization is less than 50% over all replicas for an hour, node counts will scale down. All a user has to do is define the top and bottom tiers that trigger autoscaling, and the system will handle the rest, ensuring there are no cost surprises.
“With other clouds, costs tend to only go one way, up,” said Jags Ramnarayan, SVP and SkySQL general manager at MariaDB plc. “With SkySQL, we also let you shrink the cost footprint automatically when demand is low.”
Along with autoscaling, the company is adding to SkySQL a serverless analytics layer powered by Apache Spark SQL. The move, MariaDB says, will enable companies to perform operational analytics on active transactional data as well as on external data sources, without requiring ETL. This will remove any inconsistencies between an analytical view and a transactional view.
Plus, when not conducting analytics, the service can be spun down to zero — further saving money.
John Hundley, principal software engineer at Hughes Network Systems, who got early access to the latest SkySQL release, commended the update’s user interface and autoscaling functions.
“Our IoT smart power plugs are distributed nationally across hundreds of locations, collecting data from various plugs at any given time,” he noted. “The number of locations and data rates can vary significantly. The [MariaDB SkySQL] user interface is very easy to use and will give us a better view of our database usage. We also expect autoscaling to help us in responding to our workload changes to ensure we have the right resources allocated.”
MariaDB counts organizations such as Bandwidth, DigiCert, InfoArmor, Oppenheimer, Samsung, SelectQuote and SpendHQ among its customers and competes with Microsoft SQL Server, IBM Db2 and PostgreSQL as well as the above-mentioned MySQL RDBMS.
