• Cloud Database Insider
  • Posts
  • Azure HorizonDB rivals Aurora and AlloyDB💥|Netflix boosts DB☁️|Snowflake acquires Select Star🚀|Top Vector DBs Spotlighted🔍🌐|Check out HorizonDB

Azure HorizonDB rivals Aurora and AlloyDB💥|Netflix boosts DB☁️|Snowflake acquires Select Star🚀|Top Vector DBs Spotlighted🔍🌐|Check out HorizonDB

It's not everyday that Microsoft intros a new database...Check out HorizonDB

Author

Gladstone Benjamin
December 01, 2025

In partnership with

What’s in today’s newsletter:

Also, check out the weekly Deep Dive - HorizonDB and its competitors.

⚡ EDITOR’S NOTE: I usually don’t add personal commentary at the top of the newsletter, but I think that this is a very important new offering from Microsoft. HorizonDB is a direct competitor to AWS Aurora and Google AlloyDB. Its disaggregated architecture is impressive and the fact that it incorporates AI/Vector database features. I go into detail below - don’t miss it.

Earn a master's in AI for under $2,500

AI skills aren’t optional anymore—they’re a requirement for staying competitive. Now you can earn a Master of Science in Artificial Intelligence, delivered by the Udacity Institute of AI and Technology and awarded by Woolf, an accredited higher education institution.

During Black Friday, you can lock in the savings to earn this fully accredited master’s degree for less than $2,500. Build deep expertise in modern AI, machine learning, generative models, and production deployment—on your own schedule, with real projects that prove your skills.

This offer won’t last, and it’s the most affordable way to get graduate-level training that actually moves your career forward.

RELATIONAL DATABASE

TL;DR: Microsoft launched Azure HorizonDB, a PostgreSQL-compatible cloud database offering strong consistency, high availability, and seamless Azure integration to challenge Amazon Aurora and Google AlloyDB in performance and scalability.

  • Microsoft launched Azure HorizonDB, a cloud database service competing with Amazon Aurora and Google AlloyDB.

  • HorizonDB offers high availability, strong transactional consistency, and full compatibility with PostgreSQL.

  • The service integrates tightly with Azure ecosystem and developer tools for seamless migration and improved efficiency.

  • Microsoft aims to challenge market leaders by enhancing performance, scalability, and automatic scaling features in cloud databases.

Why this matters: Azure HorizonDB strengthens Microsoft's cloud database position against Amazon and Google, offering enhanced PostgreSQL compatibility, performance, and seamless Azure integration. This intensifies competition, driving innovation and providing developers more efficient, scalable options in managed relational databases across major cloud platforms.

AWS

TL;DR: Netflix consolidated its databases on Amazon Aurora, boosting performance by 75%, enhancing scalability, simplifying operations, and improving streaming responsiveness, showcasing enterprise confidence in cost-effective, cloud-native database solutions.

  • Netflix migrated multiple relational databases to Amazon Aurora, streamlining infrastructure and reducing complexity.

  • Amazon Aurora improved Netflix's database performance by up to 75%, enhancing scalability and reliability.

  • The consolidation supports faster, more responsive streaming services, boosting user experience at large scale.

  • Netflix's move highlights increasing enterprise trust in cloud-native databases for cost-effective, high-performance solutions.

Why this matters: Netflix's move to Amazon Aurora significantly boosts database performance and scalability, directly improving streaming responsiveness and user satisfaction. This shift reduces costs and operational complexity, demonstrating a major enterprise trend toward cloud-native solutions that balance high performance with efficiency, crucial for sustaining growth and innovation in digital services.

SNOWFLAKE

TL;DR: Snowflake plans to acquire Select Star to enhance its Horizon Data Catalog with automated metadata discovery and machine learning, improving data governance, lineage tracking, and accelerating cloud analytics for enterprises.

  • Snowflake will acquire Select Star to enhance its Horizon Data Catalog for better data asset management.

  • Select Star uses machine learning for automated metadata discovery, classification, and data lineage tracking.

  • Integration will provide Snowflake users richer dataset context and faster data exploration capabilities.

  • This acquisition strengthens Snowflake’s data governance, compliance, and cloud analytics offerings for enterprises.

Why this matters: Snowflake’s acquisition of Select Star boosts its Horizon Catalog with AI-driven metadata management, improving data visibility, governance, and compliance. This advancement empowers enterprises to navigate complex data environments more efficiently, supporting better decision-making and accelerating innovation in cloud data analytics.

VECTOR DATABASE

TL;DR: GigaOm's Radar V3 evaluates vector databases, highlighting new competitors, enhanced scalability, and AI integrations, guiding enterprises to choose mature, feature-rich solutions for embedding and similarity search workloads.

  • GigaOm’s Radar V3 evaluates vector databases based on innovation, features, ecosystem support, and usability for AI applications.

  • The report highlights new entrants and expanded capabilities focusing on scalability, real-time processing, and AI workflow integration.

  • Vendors differentiate via open-source models or proprietary optimizations to address enterprise demands and hybrid query support.

  • The Radar guides organizations in selecting mature vector databases critical for AI embedding and similarity search workloads.

Why this matters: The evolving vector database market is crucial for enterprises leveraging AI embeddings and similarity searches. GigaOm’s Radar V3 offers key insights to navigate vendor choices, emphasizing innovation, scalability, and integration, which directly impact AI application performance and long-term strategic technology decisions.

EVERYTHING ELSE IN CLOUD DATABASES

DEEP DIVE

HorizonDB and its competitors.

Its not everyday that Microsoft releases a new database platform into the Azure Ecosystem. What was released was HorizonDB which is their managed PostgreSQL offering.

I think it was announced during one of their events in mid-November, but I did not find out about it until a few days ago while putting this weeks edition of the newsletter together.

So now I have to keep track of another database system.

That’s fine with me as it adds fodder to keeping this newsletter going.

Now what I have discovered is that HorizonDB is a direct competitor to AWS Aurora and AlloyDB. Even more fancy databases, based on PostgreSQL from the top 3 hyperscalers (you have to remember, I’m from the days of SQL Server 6.5).

Here is a table comparing HorizonDB with Aurora and AlloyDB, plus good OLD Postgres:

Feature Category

Azure HorizonDB

AWS Aurora PostgreSQL

Google AlloyDB

Vanilla PostgreSQL

Architecture

Disaggregated (Shared Storage)

Disaggregated (Log-Structured)

Disaggregated + Columnar Engine

Monolithic (Coupled)

Storage Scaling

Auto-scaling to 128 TB

Auto-scaling to 128 TB

Virtual Unlimited

Limited by Disk Size (typ. 64TB)

Compute Scaling

3,072 vCores (1 Primary + 15 Replicas)

~128 vCores (Limitless pending)

~128 vCores (Vertical Scale)

Single Node Vertical Limit

Vector Engine

DiskANN (SSD-Optimized Graph)

pgvector (HNSW - RAM heavy)

pgvector + ScaNN (Optimized)

pgvector (HNSW/IVFFlat)

Filtered Search

Predicate Pushdown (Vamana)

Standard Post-Filtering

Standard/ScaNN Optimizations

Standard Post-Filtering

Analytics (OLAP)

Fabric Mirroring (Parquet/OneLake)

Redshift Zero-ETL Integration

Columnar Engine (In-Memory HTAP)

Row-Store (Slow Analytics)

Caching

Tiered (RAM + Local NVMe)

Buffer Pool + Distributed Storage

Tiered (Ultra-fast Local Cache)

OS Page Cache + Shared Buffers

Availability

Multi-Zone by Default

Multi-Zone by Default

Multi-Zone (Regional)

Manual Replication Setup

Ecosystem

Fabric / Power BI / Entra ID

AWS Glue / Redshift / IAM

Vertex AI / BigQuery / IAM

Agnostic / Manual Integration

Just doing some cursory research on HorizonDB in the last couple of days, I found the scaling capabilities to be the most impressive. There are a other features that that make up HorizonDB:

  • Handling workloads of up to 128 TB

  • Scaling to 3072 vCores

  • Advanced filtered vector search

  • a curated set of pre-provisioned AI models

There is a lot of information around its release and you can check out some writings from Microsoft here and here. One other thing. As of the time of me writing this newsletter, it is only available in the Central US, West US3, UK South and Australia East regions.

One other thing before I go. You can ask Microsoft for access if you want to test out HorizonDB before GA by filling out the Azure HorizonDB Private Preview Participation Form.

Gladstone Benjamin