Highlights:

  • Although AlloyDB isn’t open-source, Google says it is fully compatible with open-source generative AI technologies, including PostgreSQL. It was created especially to deal with workloads related to AI.
  • The first database from a third party that is directly connected with VMware Cloud Foundation via VMware Data Services Manager and VMware vSAN is Google’s DBMS.

With the announcement of their partnership, VMware Inc. and Google LLC will offer PostgreSQL customers an alternative to the well-known open-source platform explicitly designed for artificial intelligence development: Google Cloud’s AlloyDB Omni database management system (DBMS) on VMware’s cloud.

In Stack Exchange Inc.’s Stack Overflow 2023 Developer Survey, respondents chose PostgreSQL as their most respected database out of 32 contenders, placing it as the fourth most popular DBMS according to DB-Engines rating by SolidIT Consulting and Software Development gmbh. Among other things, it is renowned for its high degree of SQL conformance, extensibility, and support for sophisticated features.

Although AlloyDB isn’t open-source, Google says it is fully compatible with open-source generative AI technologies, including PostgreSQL. It was created especially to deal with workloads related to AI.

Andi Gutmans, Vice President and General Manager for databases at Google, said, “Postgres is awesome, but it doesn’t have the high-end capabilities enterprises need. Think of Alloy as PostgreSQL on steroids. It’s up to two times faster on transactions and 100 times faster on analytics. It has vector search and a lot of integrations with the open-source ecosystem.”

Vertex AI is the company’s suite of tools that allows data scientists and machine learning engineers to automate, standardize, and manage machine learning projects. Users may access Google Cloud’s AI models and build vector embeddings within the database.

Transactions, Assessment, and AI

AlloyDB Omni I contains embeddings for prompt engineering in generative AI use cases and may be used for both transaction processing and analytical queries within the same database. According to VMware, VMware Data Services Manager streamlines data service management. It has procedures that interface with AlloyDB Omni to scale, replicate, patch automatically, and backup databases with little to no training requirements.

The first database from a third party that is directly connected with VMware Cloud Foundation via VMware Data Services Manager and VMware vSAN is Google’s DBMS. A VMware platform for database management and consumption is called Data Services Manager. VMware stated that it will provide administrators with a native vSphere hypervisor experience while administering AlloyDB Omni on Cloud Foundation, allowing clients to deploy database-as-a-service on-premises.

According to the company, the Data Services Manager will include embedded processes that interface with AlloyDB Omni to manage scaling, replication, backups, and database patching with little to no training required. Developers can quickly create, consume, and scale database instances due to user self-service features.

AlloyDB, which delivers most of the software through the cloud, is a rare on-premises Google offering. Gutmans said, “Customers said if they can only get AlloyDB on GCP, it’s going to make migration difficult. So, we built Omni so customers can run it anywhere with all the performance, scalability, and AI benefits.” Last month, the software became generally available.

Gutmans mentioned that Google has not decided to open-source the database so it could integrate proprietary technology like an analytics accelerator and assist with high-dimensional vector data, largely used in AI training. Still, “It always will be 100% compatible” with PostgreSQL. He added, “We made a commitment to customers that we will not be the next source of vendor lock-in.” Google is availing its cloud AI models to the omni users of AlloyDB and will also offer native models to customers with no cloud access.