While creating AI agents its very easy to use starburst it already has most support and nice number of features. If you want to let AI agents use your data there are built in plugins controlling governance and ease. Review collected by and hosted on G2.com.

Lack of community support and reliability. Most of the companies prefer to go to Azure and have their own AI solution because main thing is about reliabilty. Review collected by and hosted on G2.com.

Starburst improve the response time and incorporated AI/ML capability. Review collected by and hosted on G2.com.

My previous comment still stands. In addition, while Starburst has improved response time, certain connectors, like SQL Server, is single-threaded so the performance is lacking in comparison to other connectors that can do parallelization. Review collected by and hosted on G2.com.

Starburst has the speed of serving complex queries. I’ve found that it optimizes query execution whether the data is in multiple systems. And it helped reduce the hassle of asking people for data and querying it manually, which had skewed to a massive part of our time. Having the ability to take complex join queries, without worrying about movement of data has really elevated our workflow to be able to see more analysis instead of thinking about logistics. Review collected by and hosted on G2.com.

To be fair, Starburst is a very full featured product and pretty intuitive but some things are not so intuitive as I’d like them to be so troubleshooting can be more difficult. Moreover, when it comes to distributed query engines, those teams that are new to them will need some time to really get a whole picture of all the choices here offered. Occasionally, on a problem, I get so lost in the docs and have to dig it that we can’t work. This could certainly be made clearer and more accessible. Review collected by and hosted on G2.com.

The best thing about Starburst is how it manages such complicated data environments. It lets us query data in multiple system data, AWS and google cloud without having to worry about integration issues. It’s also fast, and that worked even with large amounts of data, which is the performance, and the results. Making our data access just much easier and indispensable for our analytics team. Review collected by and hosted on G2.com.

For some reason, the initial setup for Starburst is the main downside I’ve found to it. It’s strong, but a lot of configuration is needed when combining with other data sources. Assuring that everything is attached and working in concert can take some time. Furthermore, there are some issues with dealing with bring complex queries, which can cause a performance hiccup when trying to pull all the data from multiple sources at once. Review collected by and hosted on G2.com.

Starburst provides a single point of access to multiple disparate data sources including data lakes, cloud storage, relational databases, and NoSQL systems. This significantly simplifies the analytics process by allowing users to query data in-place, eliminating the need for complex ETL pipelines.

The Security and Governance features are great which helps in auditing etc.

1. Unified Access to Distributed Data

Starburst provides seamless connectivity to a wide range of data sources—relational databases, data lakes, cloud storage, and more—allowing analysts and data scientists to query across silos using standard SQL.

2. High Performance & Scalability

Built on Trino’s distributed SQL engine, Starburst offers fast query performance at scale. Features like cost-based optimization, dynamic filtering, and query caching significantly enhance performance for large datasets.

3. Enterprise-Ready Security & Governance

Starburst integrates with authentication systems and supports fine-grained access control via tools like Apache Ranger, making it a secure option for highly regulated environments.

4. Flexible Deployment

Supports hybrid, multi-cloud, and on-prem deployments. With Starburst Galaxy, the fully managed SaaS offering, users can get started quickly without infrastructure overhead.

5. Broad Ecosystem Support

It integrates with major cloud platforms (AWS, Azure, GCP) and connects with popular BI and data tools such as Tableau, Power BI, and dbt. Review collected by and hosted on G2.com.

While Starburst is a strong platform, there are a few areas that could be enhanced:

1. Catalog Creation and Management

Setting up catalogs, especially at scale, can be complex and sometimes unintuitive. Improvements to the user experience, automation, and management of catalogs—particularly in large, dynamic environments—would greatly benefit data engineering teams.

2. Learning Curve for New Users

Though SQL-based, the platform requires understanding of distributed query execution, connector configurations, and performance tuning. Organizations may need to invest in training or initial consulting support.

3. Monitoring and Troubleshooting

While Starburst provides basic query monitoring tools, more advanced observability features (e.g., deeper lineage tracking, proactive performance insights) could further simplify troubleshooting and operational efficiency.

4. Cost Management in Cloud Environments

Given the platform’s power, it’s easy to incur high compute costs when querying large datasets across multiple cloud sources. Resource management policies need to be carefully implemented. Review collected by and hosted on G2.com.

Starburst was one of the first to offer Fault Tolerant Mode(FTE) with Trino and it enabled us to execute larger queries without any time limit as imposed by other similar query engines like Athena.

Starburst SSO integration and SCIM capabilities along with the IAM integration with AWS was very well implemented. RBAC and ABAC capabilities are also robust.

Starburst support team has also been very responsive so far. Review collected by and hosted on G2.com.

Telemetry capabilities are still far from robust. CPU based autoscaling had a lot of issues, but has greatly improved recently. Lack of integration with alerting tools like slack or pagerduty is very limiting and making users to spend time integrating custom plugins. There is also no current feature to warn users if certain queries are blocking the cluster resources or some automatic termination of such queries. Review collected by and hosted on G2.com.

Interface

Integrations

Role base acess control (granular)

Customer Succes Review collected by and hosted on G2.com.

It would be better if AI agents operated independently from the data products entity Review collected by and hosted on G2.com.

Probably the biggest standout from Starburst is that it decreases our dependency on operator data retrieval. We no longer have to extract and put things together from separate locations because it is accessible in a structured format. It has substantially reduced redundant work and made sure that our team dedicates its time to meaningful analysis rather than an excess on data preparation. This has helped us structure our decision making a lot more and has made it more efficient. Review collected by and hosted on G2.com.

Starburst is powerful, but some advanced query functions are as complex as you would expect for something that involves moving notoriously complex data. Not all team memebrs are comfortable with some features sometimes which sometimes causes to hindrance as we wait for someone who is more experienced. The documentation could also be clearer, most especially when trying to identify a particular scenario. Because of this lack of clarity, some processes have been slower than they should be. Review collected by and hosted on G2.com.

One of the highlights of the Starburst architecture is its data lakehouse architecture. This enables us to query data between different formats and locations without any issues. This has been a lifesaver for our data analysts so that they can have faster access to and analysis of data. The cost based optimization feature is also good since it helps us manage and optimize query costs. Moreover, it is easy to integrate with different data sources for the purpose of data consolidation from different platforms, offering a single view. Review collected by and hosted on G2.com.

Setting up Starburst is a bit challenging, especially for those who are new to data lakehouse concepts. Although the documentation is comprehensive, the huge beginner friendliness. In addition, it may take a while to learn the curve of optimizing queries, and you can certainly need some trial and error to achieve the best performance. Review collected by and hosted on G2.com.

Starburst’s query engine is very powerful. This allows me to do such complex queries on mutiply data sources without lifting a finger. This is a game changer for my work with my data analysis tasks. It is impressive that the software handles large datasets efficiently. For my work, I can process and analyze lots of data but will never get a lag. Furthermore, the user friendly interface of the website makes it easy for the user to navigate and use even if one is not so tech savvy. Top notch with regards to integration capabilities as well. Review collected by and hosted on G2.com.

Starburst can handle robustly, but for beginners, it is steep learning curve. On this matter, the documentation is very complete, but for new users it can be quite overwhelming. It is also a bit confusing in pricing structure as different tiers come with different access to the features. A standard price model would be really good. Finally, the customer service is too slow to respond. Sometimes I had to wait for a longer time than I was expecting to receive help. Review collected by and hosted on G2.com.