Microsoft Azure HorizonDB Reviews

My main use case for Microsoft Azure HorizonDB is to develop multi-model services that are distributed globally to handle low latency data storage and event-driven microservice communication. I typically choose it when processing high volume or schema-evolving data such as user activity logs, audit trails, and workflow states that must be available across multiple regions for resilience and faster read and write access. We are using Microsoft Azure HorizonDB in our current project.

We are using Microsoft Azure HorizonDB as a durable event store for a distributed microservice platform where the core idea is to write every state change, whether it's a user activity event, audit log, or workflow transaction, into a single Cosmos container using the ticket ID or tenant ID as a partition key. Microsoft Azure HorizonDB automatically partitions the data and replicates it across configured regions, allowing each service to read and write with very low millisecond latency regardless of geographic location. Key components include Core SQL API for flexible JSON documents and managed identities for secure access. It auto-scales the required units, enabling us to handle traffic spikes during heavy system loads and helps us monitor using Azure Monitor alerts to catch any throttling before it impacts user experience.

The primary pattern using Microsoft Azure HorizonDB as a durable event store for our distributed microservice covers most of our needs. We also leverage complementary capabilities like enabling the change feed on the container, allowing downstream services to react to state transitions in near real-time without polling. Azure Functions are also subscribed to the feed, writing audit entries into ELK pipelines and updating materialized views in Azure Synapse for reporting, creating a clean event-driven pipeline while maintaining core service logic. Additionally, we can use point-in-time restore for multi-region write conflict resolution to guarantee data integrity when two replicas try to update the same ticket simultaneously, which is crucial for active-active deployment across NA and EMEA. For read-heavy dashboards, we use Azure Cache for Redis to cache the most recent ticket snapshots, pulling them from Cosmos only when change feed signals an update. Overall, these extensions allow us to transform Microsoft Azure HorizonDB from a durable store into a reactive low-latency backbone that supports real-time processing and analytical workflows, while ensuring high availability and cost efficiency for our project.

The biggest positive impact of Microsoft Azure HorizonDB on my organization is its transformation of our event-driven microservice platform into a globally low-latency, highly available data backbone. By persisting every state change, such as a ticket moving from open to in progress, into a single Cosmos container using the ticket ID as a partition key, we achieve automatic horizontal scaling with sub-millisecond read/write latencies across different regions. Overall, Microsoft Azure HorizonDB serves as a durable, flexible schema store that can scale transparently, improving reliability for enterprise-level software, allowing us to focus on business logic rather than infrastructure configuration, thus delivering a faster user experience with lower operational overhead for the team.

One of the biggest impacts we have experienced from moving to Microsoft Azure HorizonDB is the combination of global low latency read and write capabilities. By persisting average state changes, such as moving a ticket to progress, it handles requests in milliseconds, while Microsoft Azure HorizonDB manages multi-region replication and conflict resolution with auto-scaling request units. As a result, we eliminated the need for a custom sync layer, raising our system availability to 99.95% during regional outages. The change feed, which feeds into Azure Functions, cuts our mean time to resolution by over half, leading to a 38% reduction in API response time. This translates directly to faster user experiences, and the auto-scaling helps lower our monthly operational spending by roughly 30% compared to previous self-managed database clusters.

The standout features of Microsoft Azure HorizonDB are its native multi-model support and global distribution with turnkey scaling, along with rich consistency and change feed capabilities. It natively handles document, key-value, graph, and column family data, allowing me to pick the API that matches the workload such as Core SQL, flexible JSON, or MongoDB for existing drivers. It also automatically partitions data across physical nodes and replicates it to any number of regions I configure, providing sub-millisecond latency for both read and write operations globally. The operational features like point-in-time restore for multi-region write conflict resolution and serverless auto-scaling of request units keep the system highly available and cost predictable. Integration of security via managed identities and role-based access control removes the need for secret management, while Azure monitor metrics let us tune the system before throttling occurs. In practice, I utilize these features to build an event-driven microservice platform where every state change is persisted in Microsoft Azure HorizonDB and instantly propagated via change feed through Azure Functions, along with auditing and analytics cached using Azure Redis for ultra-fast dashboard reads. Overall, the benefits provide a relatable low-latency, globally resilient data backbone, and it can automatically scale while minimizing operational overhead for developers.

The feature that has the most impact on my work is the built-in change feed. It allows microservices to write a state change to Microsoft Azure HorizonDB and allows downstream consumers to react automatically without polling or using custom synchronization code. The server-side feed scales with throughput and guarantees in-order, nearly real-time delivery for every message. I pair it with Azure Functions to either write audit records for the ELK pipeline or update materialized views in Azure Synapse. This integration helps decouple our core services from analytical and notification layers, dramatically reducing latency. For example, when a ticket moves through the workflow, the change feed triggers a function that both persists the transaction for reporting and invalidates the Redis cache for the UI, resulting in very low latency and end-to-end visibility. It is simpler than the previous code base, and the overall change feed provides real-time relativity, leading to lower operational overhead and scalability for our event-driven architecture.

The feature that has given me the biggest productivity boost is the built-in change feed, which I use for persistent changes. For example, as a ticket moves from open to in progress, downstream consumers react automatically without the need for polling or custom sync code. This pattern provides built-in durability, and we can configure multi-region write conflict resolution, keeping the system highly available during regional outages. In practice, this gives us sub-millisecond end-to-end visibility, which is much simpler than before.

From a tooling perspective, Microsoft Azure HorizonDB provides built-in support for Azure OpenAI embeddings, allowing us to generate vectors on the fly in serverless functions and write them back into the same container. Managed indices automatically scale our request units while maintaining the same multi-region replication and consistency guarantees, keeping latency low and the system highly available. In practice, I have used embeddings in our container to power a real-time FAQ bot, where each incoming question receives embeddings and triggers a function that queries the vector index, producing results for users within milliseconds. This setup eliminates the need for a separate search microservice and streamlines our architecture. Overall, the AI extension adds significant value by simplifying our embedding workloads while maintaining Microsoft Azure HorizonDB's reliability.

I find the AI output from Microsoft Azure HorizonDB to be both accurate and reliable for most of our use cases. It consistently meets expectations as the service uses OpenAI embeddings with multi-region replication guarantees that we trust for transactional data. The vector index delivers results with millisecond latency, and in practice, we built a FAQ that shows relevance scores comparable to a dedicated search service, enabling users to see the correct answer about 92% of the time. Occasional mismatches typically stem from the underlying language model rather than the storage layer itself. The change feed-driven pipelines ensure that new or updated content is indexed almost instantly, keeping the model fresh with no need for re-indexing. Overall, these extensions have met my expectations for accuracy and reliability, delivering fast and dependable embedding search while keeping the architecture simple and highly available.

I would rate Microsoft Azure HorizonDB a 9 out of 10 because it is a solid service provided by Microsoft that reduces API latency by 38% and improves system availability to 99.95%. The only reason I do not rate it a perfect 10 is that I would prefer to see richer server-side query features and more granular cost forecasting in the portal, which would streamline development and debugging tasks. The overall trade-offs are minimal compared to the reliability and performance gains it offers. A few enhancements around observability would further strengthen its differentiation among competitors.

I have been using Microsoft Azure HorizonDB for roughly around three years, starting in early 2022. I began by integrating Azure DevOps pipelines, and we have extensively used Azure Blob Storage and Microsoft Azure HorizonDB in our project. I have around two and a half years of experience in Microsoft Azure HorizonDB, maintaining data and scaling it.

Microsoft Azure HorizonDB is very stable in my experience.

Microsoft Azure HorizonDB is auto-scalable and automatically handles everything needed for database scaling. Microsoft Azure HorizonDB does an excellent job with the auto-scaling part.

Customer support for Microsoft Azure HorizonDB is excellent and available 24/7, and I would rate it a 10 out of 10 for effectively addressing all customer situations.

We experience a significant return on investment with Microsoft Azure HorizonDB, as it leverages multi-region and active-active replication through auto-scaling. We have seen major improvements in our response time, achieving 99.9% reliability. Additionally, it reduces the need for human attention and lowers overhead related to infrastructure configurations, resulting in considerable benefits.

I find the pricing for Microsoft Azure HorizonDB very fair and competitive, and the setup cost has sufficient documentation available for newcomers. They provide comprehensive support, responding immediately to issues 24/7, leaving me with no questions concerning pricing, support, or documentation since everything is quite good.

I suggest everyone give Microsoft Azure HorizonDB a chance to see the improvements and cost-effectiveness it offers for current production systems.

I use Microsoft Azure HorizonDB for groups where transactional workloads are high, for PostgreSQL applications. For example, I need to scale compute and storage as and when the demand changes, and that's where I use Horizon database.

Microsoft Azure HorizonDB's machine learning capabilities assist my organization's data analytics efforts because it helps with PostgreSQL extensions such as PGVector that are useful for retrieval-based AI applications. In my rack-based applications, ML models require very fast and relevant access to data, and that's where it helps.

I decided to go with Microsoft Azure HorizonDB because it's new and, as an AI-native company, we always want to try new things. I expected some ROI and efficiency improvement, so I decided to adopt it because I wanted a scalable and high-performance database that supports my AI use cases.

I value Microsoft Azure HorizonDB because it provides strong PostgreSQL compatibility, and that too with scalable compute and storage. Whenever any workload scales or whenever the traffic increases, it doesn't require manual management of the underlying database infrastructure.

Microsoft Azure HorizonDB's tools for automated backups are excellent because I don't have any constraints, and I can keep secure database backups that are always available.

Microsoft Azure HorizonDB's versatile data models help in fulfilling my applications' diverse requirements by providing PostgreSQL data models that support structured data. I use tables, rows, and columns for SQL, which makes it suitable for my AI use cases such as semantic searches based on those requirements.

Microsoft Azure HorizonDB's monitoring features have improved database performance significantly since I don't have to do it manually.

The initial setup process for Microsoft Azure HorizonDB was very straightforward because it was through the Azure environment. I created database clusters and then configured it, taking two to three days for production setup testing, which is doable for anyone.

I think Microsoft Azure HorizonDB can improve by maturing its features, especially since the product is pretty new right now. More AI/ML tooling and integrations will make adoption easier, and those features would significantly help.

While the functionality of Microsoft Azure HorizonDB is good, I would like to see more third-party integrations in the future, as it is relatively new compared to others. With time, it will gain more production maturity and instill more confidence in users, but right now, improvements in AI/ML tooling and integration are what I feel are necessary.

I have been working with Microsoft Azure HorizonDB for four months now, and this is something new for my organization.

Microsoft Azure HorizonDB helps to streamline my workflows because it has favorable compatibilities to keep applications available as and when required. My downtime is very minimal, allowing my workflows to work effectively without much interruption.

Using Microsoft Azure HorizonDB provides better scalability and performance in terms of database administration. Instead of manually provisioning additional databases to infrastructures whenever workload increases, I can scale the database environment based on application requirements with the help of Horizon database environment.

It basically reduces the infrastructure management overhead, and also helps in scaling the storage and compute independently, so I don't have to manually do it. This feature matches well with my workload requirements.

Through my Azure ecosystem prior to adopting Microsoft Azure HorizonDB, I basically had data privacy and security controls.

The initial setup process for Microsoft Azure HorizonDB was very straightforward because it was through the Azure environment. I created database clusters and then configured it, taking two to three days for production setup testing, which is doable for anyone.

So far, I have not seen any return on investment from Microsoft Azure HorizonDB in terms of cost. However, I realize that it can improve, and I have sensed that it reduces infrastructure management time, allowing for quick storage scaling as per requirements, which I consider the major ROI, although it hasn't been properly measured yet.

I don't have any idea about the pricing aspect of Microsoft Azure HorizonDB, but I have heard in meetings that it's relatively costly.

My advice for other organizations considering Microsoft Azure HorizonDB is that anyone who wants to try it should first test it in a controlled workload before moving everything to Horizon. It's important to focus on data modeling before going live and to evaluate the performance properly before planning full implementation, especially for those needing a new solution with ML workloads. I would rate this product an eight out of ten.

My main use case for Microsoft Azure HorizonDB is that we were experiencing performance issues in Azure PostgreSQL flexible server, so we wanted to eliminate those issues and started using Microsoft Azure HorizonDB. A specific example of how I used Microsoft Azure HorizonDB to address those performance issues is that we were experiencing disk IO errors in Azure PostgreSQL. Microsoft Azure HorizonDB has a separate storage layer, so when we started using it, we did not encounter those disk IO errors. Regarding my main use case and how it has helped with those disk IO issues, we did not perform any special configuration in Microsoft Azure HorizonDB. The underlying architecture itself has a separate storage layer, whereas Azure PostgreSQL shares CPU, memory, and storage layers together. We believe this different architecture may have helped resolve the performance issues. Because our application is very disk IO intensive with heavy read and write operations, this is one of our critical databases.

The best features Microsoft Azure HorizonDB offers is a separate storage layer, which I feel is the best feature. We also have noticed a few security improvements compared to Azure Database for PostgreSQL. The security improvements I noticed in Microsoft Azure HorizonDB compared to Azure Database for PostgreSQL stand out to me. When integrating it with our Azure Entra, we faced difficulties with Azure Database for PostgreSQL, but in Microsoft Azure HorizonDB, the setup was very simple for configuring the integration with our Entra. We also have centralized access control alongside the rest of our Azure identity estate, which we do not have with PostgreSQL. Microsoft Azure HorizonDB has impacted my organization positively, although it is not in full-fledged usage yet. We just started what we can say is a proof of concept setup and it is not in actual live usage yet. Since we are in the proof of concept phase, we have noticed some early improvements and potential benefits that stand out so far in terms of speed, reliability, and ease of management. The performance is very much improved. The security posture including Entra ID, private endpoint, encryption, and zone replication also shows some improvement compared to our existing Azure PostgreSQL flexible servers.

Improvements for Microsoft Azure HorizonDB mean that we still have not gotten actual usage of real issues. We just started the proof of concept phase, so we might discover what the exact real issues are once we get full-fledged usage at the ground level. Regarding needed improvements, I feel that we need more documentation on Microsoft Azure HorizonDB. The documentation is less comprehensive when compared to existing Azure managed SQL instances, PostgreSQL, and MySQL. We still require more documentation on Microsoft Azure HorizonDB. We also do not have detailed architecture diagrams the way we have for managed instance, PostgreSQL, and MySQL. The documentation needs to be improved.

I have been using Microsoft Azure HorizonDB for three months.

Microsoft Azure HorizonDB is stable.

Microsoft Azure HorizonDB's scalability is good. One more feature improvement I observed compared to our Azure PostgreSQL PaaS instance is scalability, which is good in Microsoft Azure HorizonDB.

We have Azure overall support for customer support.

Previously, we were using AWS Aurora PostgreSQL, which we feel is a little bit pricier compared to Microsoft Azure HorizonDB. All our applications and setup are mostly Microsoft user-friendly applications, so that is why we started using Microsoft Azure HorizonDB.

When integrating Microsoft Azure HorizonDB with our Azure Entra, we did not face difficulties. The setup was very simple for configuring the integration with our Entra. We also have centralized access control alongside the rest of our Azure identity estate.

I have not seen a return on investment yet, as we have still not finalized or determined what the final ROI will be.

My experience with pricing, setup cost, and licensing for Microsoft Azure HorizonDB is that the pricing is good.

Before choosing Microsoft Azure HorizonDB, we evaluated AWS Aurora PostgreSQL.

On a scale of one to ten, I would rate Microsoft Azure HorizonDB overall as nine out of ten. I chose nine out of ten because of the documentation and the real use cases that I still have not seen published by Microsoft. Regarding Microsoft Azure HorizonDB's AI capabilities, I still have not covered the governance and security aspects, including what guardrails or security mitigations we need to have in place for Microsoft Azure HorizonDB. I think the security enhancements that we are doing for PostgreSQL, MySQL, and managed SQL instances will apply here, but I still have not gone into detail about any specific security requirements we need to implement for Microsoft Azure HorizonDB. Regarding Microsoft Azure HorizonDB's AI capabilities, I feel that its accuracy and reliability of output is good. I feel that Microsoft Azure HorizonDB intelligence might resolve performance issues or in-built capabilities such as re-indexing and database object tuning. I feel that it has good AI capability compared to our other Azure PaaS instances. There are no other improvements needed for Microsoft Azure HorizonDB that I have not mentioned yet. Something small that I hope to see in the future would be the documentation and real use cases. My advice to others looking into using Microsoft Azure HorizonDB is that instead of going with open source, I recommend using Microsoft Azure HorizonDB. I do not have any additional thoughts about Microsoft Azure HorizonDB before we wrap up, as I feel I have already provided all the inputs I have.

Microsoft Azure HorizonDB serves as my main use case because it is fundamentally a PostgreSQL database, and our company manages financial data with this type of database. It functions as an OLTP system, and with workflows involving massive data, we manage it using Microsoft Azure HorizonDB.

A specific example of how I use Microsoft Azure HorizonDB with our financial data is that it contains multiple datasets and is a huge database. It operates as an OLTP, and when multiple users connect simultaneously, we found that it manages the transaction volume very easily and very smoothly. The application users are not facing any deadlock or locks in this scenario, and it is also used for real-time analytics very smoothly. It has outstanding performance compared to other relational databases, which is why we are using it. Microsoft Azure HorizonDB is better than other relational databases.

The best features of Microsoft Azure HorizonDB that I find valuable are its tremendous performance tuning benefits. First, it eliminates the checkpoint bottlenecks. Traditional PG Flex databases write dirty data pages from memory to disk, causing massive IOPS bottlenecks, but Microsoft Azure HorizonDB has different mechanisms. It streams only the WAL to the storage nodes where the page state is reconstructed asynchronously, which reduces the consistency and the commit latency is very low.

From the backup perspective, one of the features of Microsoft Azure HorizonDB that I would highlight is its zero impact snapshot backups, which I believe are better than the Azure PG Flex database. It relies on instantaneous storage blob snapshots, which is excellent, and continuous WAL archiving. I believe the recovery point is also better compared to PG Flex, which is where it excels.

Microsoft Azure HorizonDB has positively impacted my organization by reducing latency issues, improving performance, and providing a good recovery time. These are the things that are very beneficial for my organization.

I think one excellent feature that would improve Microsoft Azure HorizonDB is similar to what Oracle has, the restore point, such as time travel. I believe that time travel using a restore point is not possible here, and that is one feature I think is missing.

I have been using Microsoft Azure HorizonDB for more than one year; specifically, 15 months.

My advice for others looking into using Microsoft Azure HorizonDB is to proceed if you want to pay more and if you want to stabilize your environment and use the AI-related features. I want to say that the costing of Microsoft Azure HorizonDB is too high and should be less. I would rate this product a 7 out of 10.

Microsoft Azure HorizonDB serves as our main use case to support production applications that handle large volumes of semi-structured and frequently changing data, and we use it as a scalable NoSQL database for application records that require fast and reliable access. For example, we use Microsoft Azure HorizonDB as the backend database for a business application that stores user profiles, sessions information, application activity, configuration data, and other operational records.

Using Microsoft Azure HorizonDB for this use case makes the application easier to manage and scale while also providing fast and reliable access to frequently used data, and the flexible NoSQL data model allows our development team to make changes to application data structures without extensive database schema modifications, which reduces the operational effort for our team because the underlying data infrastructure, availability, scaling, and routine maintenance are handled through the managed service.

One additional benefit of Microsoft Azure HorizonDB is that it fits well into our overall Azure workflow because it can be integrated with our application, monitoring, security, and analytics services, allowing the database to be managed as part of the broader cloud environment rather than as a separate infrastructure component. The flexible data model is also useful when application requirements change frequently, enabling our team to accommodate evolving data structures without significant database redesign, while the managed service reduces routine administration and infrastructure maintenance.

Microsoft Azure HorizonDB has had a positive impact on our organization by reducing the operational effort required to manage scalable NoSQL workloads and making it easier to support applications as user activity and data volumes grow. One of the main improvements we have noticed is reduced database administration, as the managed service handles many routine infrastructure and maintenance activities, allowing our team to spend more time on application development, performance optimization, and other high-priority initiatives.

Since adopting Microsoft Azure HorizonDB, we have seen measurable benefits, estimating that it saves our team around five to ten hours per month by reducing manual effort related to database administration, capacity planning, maintenance, and infrastructure management. We have also seen improvements in how quickly we can respond to changes in application demand; scaling the database environments is easier than managing additional database infrastructure manually, which reduces provisioning and operational delays.

The best features of Microsoft Azure HorizonDB in our experience are its scalability, low latency performance, flexible NoSQL data model, high availability, and managed database operations, with the ability to scale as application traffic and data volume increases being particularly valuable for our production workloads. We find the flexible document-based model useful because application data structures can evolve without requiring extensive schema changes, and another important advantage is the managed nature of the service, which reduces the amount of effort our team needs to spend on infrastructure management, routine maintenance, availability, and capacity planning.

The feature that has made the biggest difference for our team is the scalability combined with consistently low latency performance, which allows our application workloads to change based on user activity and data growth without manually provisioning and managing additional database infrastructure. From a daily operations perspective, this has reduced the amount of time our team spends on capacity planning, database maintenance, and infrastructure-related troubleshooting.

We consider the governance and security capabilities of Microsoft Azure HorizonDB in an AI-related use case to be strong, with the integration into Azure Identity and Access Management, encryption, networking, and monitoring and auditing providing a solid foundation for protecting data used by AI-enabled applications. We particularly value the ability to control access to database resources and apply existing Azure security and governance policies, which helps maintain appropriate permissions and security boundaries when AI or analytics services interact with application data.

We find the AI-related output associated with Microsoft Azure HorizonDB to be generally accurate and useful, particularly for tasks analyzing application data, identifying patterns, and assisting with query or performance-related recommendations. One of the main strengths is that the AI capabilities help reduce manual analysis and troubleshooting and development activities, with the recommendations being particularly useful as a starting point when investigating performance issues or understanding application data.

While the performance of Microsoft Azure HorizonDB is generally good, we have noticed that it can be more challenging during periods of higher application traffic and larger data volumes, with the impact being more noticeable when workloads generate a high number of concurrent read and write operations or when queries involve larger data sets, as poorly optimized or partitioned usage can result in higher resource consumption and require additional tuning.

I have been using Microsoft Azure HorizonDB for around three years, during which time we have used it for production application workloads involving scalable semi-structured data and have gained good experience with its performance.

Microsoft Azure HorizonDB is very stable in my experience.

Microsoft Azure HorizonDB has been highly scalable for our workloads, allowing us to accommodate increases in application traffic, user activity, and data volumes without manually provisioning and managing additional database infrastructure. We particularly value the ability to adjust capacity based on changing workload requirements, with the distributed architecture and flexible NoSQL data model making it suitable for applications that need to handle growing amounts of data and concurrent requests.

I have reached out to customer support for Microsoft Azure HorizonDB, and the customer support was very prompt and fast, answering all our requests and calls with prompt support.

I would rate the customer support for Microsoft Azure HorizonDB a ten.

I am using Microsoft Azure HorizonDB only.

We have seen a positive return on investment from Microsoft Azure HorizonDB, primarily through reduced operational effort and time saved rather than by reducing headcount, estimating that Microsoft Azure HorizonDB saves our team approximately five to ten hours per month by reducing manual work related to database administration, infrastructure maintenance, capacity planning, and troubleshooting. The managed service also means we do not need to spend as much time provisioning or maintaining underlying database infrastructure.

Our experience with Microsoft Azure HorizonDB's pricing, setup cost, and licensing has been generally positive, as there was no major upfront hardware or infrastructure investment due to it being a managed cloud service. The initial setup is relatively straightforward compared with deploying and managing a database environment ourselves, and the licensing model is integrated into Azure's consumption-based pricing, so costs are primarily driven by throughput, storage, and workload usage, which provides flexibility to align resources with application requirements rather than making a large upfront infrastructure investment.

We did not evaluate other options and directly chose Microsoft Azure HorizonDB.

I would rate Microsoft Azure HorizonDB overall as a ten.

I give Microsoft Azure HorizonDB a ten because it provides the combination of scalability, low latency performance, flexibility, high availability, and managed operations that our production workloads require, with what stands out most being the ability to handle growing application traffic and data volumes without managing the underlying database infrastructure ourselves, along with the flexible NoSQL data model that makes it easier to accommodate changing application requirements without extensive schema redesign.

I would recommend Microsoft Azure HorizonDB for organizations that need a scalable, low-latency NoSQL database for applications with semi-structured or rapidly changing data, and teams should carefully evaluate the data access patterns before adopting it. My overall review rating for Microsoft Azure HorizonDB is ten out of ten.