Insights from Microsoft on the New PostgreSQL Variant
In a recent discussion with Shireesh Thota, Corporate Vice President of Databases at Microsoft, the conversation centered around the introduction of a new variant of PostgreSQL tailored for the cloud environment. Thota articulated the growing trend among users who are transitioning their PostgreSQL databases from community versions hosted on-premises or in other cloud platforms to Azure. This migration is driven by a desire for a fully integrated PostgreSQL experience, complete with operational flexibility and enhanced performance.
“I think increasingly what we notice is that people either go into the bucket of, ‘I want to lift and shift my PostgreSQL that’s working in the community version on-premises, or maybe another cloud.’ They want to move it to Azure. They want 100% Postgres. They want all extensions working. They just want something that really has the flexibility of performance and speed. Then Azure Database for PostgreSQL, the existing version is perfect. Somebody who wants to build an AI-native, cloud-native kind of a workload that may need a lot of storage, wants really fast latencies, significantly higher IOPS. Then you go to HorizonDB.”
The performance metrics associated with Azure HorizonDB are noteworthy, showcasing a threefold increase in throughput compared to its open-source counterpart when handling transactional workloads. This new service can scale up to an impressive 3072 cores and offers up to 128TB of storage with sub-millisecond commit times. Such capabilities are particularly beneficial for AI applications and large-scale Kubernetes deployments. Thota emphasizes, “These cloud-native workloads can really succeed on HorizonDB.”
At the heart of this performance enhancement lies a significant architectural evolution. HorizonDB separates compute and storage, allowing each to scale independently based on user requirements. This means that if additional compute power is necessary, HorizonDB can provide it seamlessly. Similarly, if there is a need for more read replicas, the system can provision them without disruption.
