Starburst Galaxy

What do you like best about the product?

Connect everything—databases and cloud—faster. They eliminate the need to move data around. You can work with clean data and use natural language by asking simple questions, without any migration.

What do you dislike about the product?

Well, what I didn't like much is that the beginning is a bit difficult, you don't know how to unconnect and use it, not to mention that you need to know SQL-based systems. Since data streaming is very dependent on the network, if a database has a slow connection, the entire Stardust query will be slowed down.

What problems is the product solving and how is that benefiting you?

The main problem this helped me solve was the chaos of fragmented data—when information is spread across different servers and sources like AWS, Google Cloud, local databases, or even Excel. It supports real-time decision-making, boosts my productivity right away, and gives me strong control over my data.

What do you like best about the product?

I particularly appreciate the robust query capabilities and user-friendly interface of Starburst. The ability to execute complex queries against diverse data sources without writing extensive SQL code is a game-changer. For instance, I often need to analyze large datasets from multiple cloud storage systems simultaneously, and Starburst's unified catalog allows me to do this efficiently with just a few clicks. This significantly reduces the time spent on manual integration and data preparation tasks.

Additionally, the dynamic schema discovery feature automatically detects table schemas for new data sources, which is incredibly helpful when working with constantly changing datasets in my organization. This saves hours each week that would otherwise be wasted setting up and maintaining connections to different databases manually.

What do you dislike about the product?

One area for improvement is the initial setup and onboarding process, which can be quite complex and time-consuming. Configuring the unified catalog requires meticulous attention to detail and an understanding of each data source's schema and metadata. While the documentation provides a good starting point, it lacks detailed examples for more advanced configurations involving third-party systems.

Another pain point is the performance when handling extremely large datasets in real-time analytics scenarios. Although Starburst performs well under most conditions, there are occasional delays and scalability issues during peak usage times or when querying very large tables. This can impact productivity and user satisfaction, especially for time-sensitive projects.

Improving the onboarding experience through more interactive tutorials and guided setup wizards could alleviate some of these initial frustrations. Additionally, enhancing query optimization techniques and resource management features would help maintain performance even under heavy loads, ensuring a smoother user experience throughout all phases of data analysis.

What problems is the product solving and how is that benefiting you?

Before using Starburst, we struggled to integrate and analyze data from multiple, disparate sources, including cloud databases, SaaS applications, and on-premises systems. This fragmentation created inefficient workflows: team members spent a lot of time manually preparing, reconciling, and syncing datasets before any analysis could even start. Without a unified catalog, finding and accessing the most up-to-date data was cumbersome and frequently led to delays.

Since implementing Starburst, we can seamlessly query across these diverse data sources through a single interface, which has dramatically improved our overall efficiency. Instead of spending days setting up connections and preparing datasets, I can now run complex analytics queries within minutes, supported by the unified catalog and dynamic schema discovery. As a result, we’ve achieved substantial time savings—about 30 hours per week across my team—and we’re able to spend more time on generating insights rather than managing data.

In terms of measurable impact, we’ve reduced project timelines by roughly 50%, and our ability to respond to business needs with real-time analytics has improved significantly. This strengthens decision-making and helps drive better ROI by enabling faster deployment of data-driven initiatives.

What do you like best about the product?

I found Starburst very useful overall, especially because it made it easy to query data across different systems without having to move everything into one place. The experience felt practical and efficient, and it was a good fit for working with distributed data.

UI / UX
I liked the UI because it was clean and easy to use. It made common tasks feel straightforward, and I didn’t spend much time figuring out where things were.

Integrations
One of the biggest positives for me was the integrations. It connected well with multiple data sources, and that saved a lot of time in setup and day-to-day work.

Performance
Performance was a strong point. Queries felt fast, and that made analytics and reporting much smoother than I expected from a federated setup.

Pricing / ROI
From my perspective, the value came from reducing data movement and saving engineering time. Even if the pricing isn’t the cheapest, the ROI makes sense when you factor in faster access, simpler architecture, and less operational overhead.

Support / Onboarding
Onboarding was smooth, and the support was responsive when I needed help. The documentation and guidance made it easier to get up and running without a lot of friction.

AI / Intelligence
I also liked that Starburst is moving toward AI-ready use cases. The idea of combining federated data access with governance and AI-friendly features makes it feel more future-proof.

Overall, I’d describe Starburst as a strong platform for teams that want fast, federated SQL access across many systems, with good usability and solid support

What do you dislike about the product?

I’d say my biggest dislikes are the steep learning curve, the setup/configuration complexity, and the fact that it can feel expensive as usage grows. For smaller teams, that combination can make it harder to justify unless the federation and performance benefits are clearly worth it.

UI / UX
The UI is usable, but it doesn’t completely hide the complexity underneath. When a platform has a lot of configuration and governance depth, the day-to-day experience can feel heavier than simpler tools.

Integrations
Even though the integrations are a strength, the breadth of connectors can also add operational complexity. In practice, that means there can be more moving parts to manage and more places where setup needs to be exact.

Performance
Performance is generally strong, but it isn’t always perfect for every workload. Some users mention slow queries or performance issues, especially when they’re asking very specific or demanding questions.

Pricing / ROI
Pricing is probably the part I’d question most. The value can be great at scale, but for smaller teams or tighter budgets, the cost can rise quickly as query volume grows.

Support / Onboarding
Onboarding is where I’d expect the most friction. It seems like you need enough technical depth to configure things correctly from the start, otherwise the learning curve can slow the team down.

AI / Intelligence
The AI direction is promising, but I’d still treat it as emerging rather than fully proven for every use case. For me, it wouldn’t be the main reason to choose Starburst today.

What problems is the product solving and how is that benefiting you?

Starburst is mainly solving the problem of data silos: instead of copying data into one central place, it lets you query data where it already lives across warehouses, lakes, databases, and SaaS tools. That also helps with slow, fragmented analytics, because federated query, pushdown, parallelization, and cost-based optimization can make cross-source queries faster and less operationally painful.
For me, the biggest benefit is that I can get a single SQL layer over distributed systems, which means less ETL, less duplication, and less time spent stitching data together manually. It also reduces operational overhead because I don’t need to build and maintain as much custom query infrastructure just to unify sources.
In practice, that means faster access to data, easier cross-system analysis, and quicker decision-making. It’s especially valuable when I need to join data from different platforms in real time or when I want to modernize an architecture without moving everything first.
The real value is not just convenience; it’s that Starburst helps keep data in place while still making it usable for analytics and AI workflows. That gives me more flexibility, better governance, and a cleaner path to scaling analytics without constantly rebuilding pipelines.

What do you like best about the product?

It has better performance, UI/UX and support. Also we consider cost efficiency for this purchase. It has rich integration options for our use case. I will highly recommend.

What do you dislike about the product?

Licensing can get expensive as clusters scale, especially compared to running open-source Trino yourself. For teams already comfortable managing infrastructure, the premium can feel hard to justify.

What problems is the product solving and how is that benefiting you?

Starburst lets you query data across multiple sources (data lakes, warehouses, databases) through one SQL interface, without moving or duplicating everything into a single system first. That's the core problem it solves: data fragmentation across teams and platforms.

What do you like best about the product?

What I like most is that Starburst makes it easier to work with data located in different places. Instead of constantly transferring and copying data, it is possible to simply access different sources through a single interface. This way, teams can quickly get the information they need and work more efficiently with large amounts of data.

What do you dislike about the product?

What I like less about Starburst is that it takes a little time for users to get accustomed to all the features of the platform. For new users, it can seem complex at first, especially when dealing with larger amounts of data. I wish there were more simple guides and examples for everyday user scenarios.

What problems is the product solving and how is that benefiting you?

Starburst helps us to combine data from different sources and quickly access useful information. We mostly use it for reporting, data analysis, and making decisions based on up-to-date data.

What do you like best about the product?

I use Starburst to 'dismantle' our data from D365FO efficiently using BYOD, and it's great for querying large amounts of data quickly. I like how Starburst allows me to join large datasets across silos without needing custom pipelines. The separate computing from storage feature and its parallel processing capabilities help in making data processing faster than our current methods.

What do you dislike about the product?

I wish Starburst had a direct connection to D365FO because using BYOD or any middleman adds extra cost, specifically in Azure. The initial setup was a bit harder since we had to create a separate BYOD database. If a direct connection were possible, it would definitely be easier.

What problems is the product solving and how is that benefiting you?

I use Starburst to dismantle data from D365FO, enabling us to query data faster and join large datasets across silos. It separates computing from storage and works in parallel, enhancing speed.

What do you like best about the product?

Starburst has good and user friendly UI. It also provides a very good performance. The out of the box integrations saves hours of work every week.

What do you dislike about the product?

So far I haven't found any downsides of using starburst. Everything is really good. The AI Integration is really good. And it can be improved further to increase the AI coverage throughout the product.

What problems is the product solving and how is that benefiting you?

Starburst acts as a one-stop-shop tool for analytics. It helps me get meaningful insights from my data, and it takes less than a day to set up. With other tools, it typically takes at least a week to reach the same point.

What do you like best about the product?

The federated query layer — one SQL surface over S3/Iceberg and Redshift — combined with a solid public API that lets us automate the whole cluster/catalog lifecycle for self-service onboarding, on managed elastic compute we don't have to operate. The AI capabilities are an add-on advantage

What do you dislike about the product?

Currently rough edge is managing catalogs at scale (per-cluster limits and some API latency), but it's easy enough to work around.The pricing runs a little on the higher side, so cost is something we keep an eye on as usage grows

What problems is the product solving and how is that benefiting you?

We struggled with data siloed across S3/Iceberg and Redshift — requiring copies, ETL, and our own query infrastructure to unify it — but now we can query every source through one SQL layer with compute and connections provisioned programmatically, which has resulted in faster access to data, much less ETL and duplication, and significantly lower operational overhead through managed, elastic clusters.

What do you like best about the product?

Starburst solves the challenge of ingesting streaming security data via schema discovery while allowing querying and enrichment from federated catalogs at near real time and allows the collapsing of multiple applications into one product thereby reducing cost.

What do you dislike about the product?

Starbursts own application logging could be more detailed and easier to export for historically lookup and review. When a query fails and you use the advance tab to look into specific Trino errors at run time, there is additional steps to troubleshoot/discover what they suggest as the issue.

What problems is the product solving and how is that benefiting you?

The ever-expanding list of catalogs and vendor integrations creates an environment where we no longer have the burden of ingesting and storing and managing all the company's data but provides it seamlessly for our data analytics and data scientists' use cases. There is less product sprawl, less administrative overhead and simpler cross catalog analytics.

My main use case for Starburst Galaxy is querying petabytes of data across vast data sources, and I use a federated query engine to join data sources from different databases and then join them using Starburst Galaxy.

I have different data sources, including Oracle, DB2, and a MongoDB cluster, so I join all of these data sources using Starburst Galaxy with the federated querying feature. I transform that into Iceberg using Starburst Galaxy, land it in S3 storage, convert it into Iceberg tables, and then use them for dashboarding in Power BI or Tableau.

Starburst Galaxy offers me several best features, which include very fast querying results, automatic indexing of data for long tables, a cost-based optimizer which reduces the time to query large tables, and an agentic feature that lets me talk to my data.

I find myself relying most on querying from different databases as well as automatic indexing in my day-to-day work, as I am a data science architect who needs to get the queries in a very short period of time. Starburst Galaxy serves the best purpose for me because if my SLAs are not met with my customers, they will raise a case, and I have tried many other tools, but Starburst Galaxy fits the best.

Starburst Galaxy has positively impacted my organization since we were struggling with Denodo and Dremio, which had their own features but were not helpful in querying large amounts of data, especially semi-structured or unstructured data. Starburst Galaxy addresses this with many YAML files and manifest files for automated maintenance, and it helps reduce the small file problem in different HDFS systems. Additionally, Starburst Galaxy has an MCP server that connects to various agentic pipelines, reducing the time to market for data consumption.

Starburst Galaxy can be improved by discovering unstructured data and building in streaming ingestion because we are currently using Kafka for that purpose. We rely on third-party tools for ingesting the streaming files, and I see they are integrating with MCP and agentic pipelines. Including these features would make Starburst Galaxy a much better tool.

I have been using Starburst Galaxy for the last four years.

Starburst Galaxy is stable.

Starburst Galaxy's scalability is excellent as it can easily scale up to large clusters, with many nodes configured for the amount of data ingested daily, allowing it to handle petabytes of data efficiently.

Customer support is quite good; we faced issues with complex queries and reached out to Starburst, and they were very helpful in troubleshooting those issues.

We previously used Dremio, but it was not fast and had a lot of limitations in processing large unstructured and structured data, leading us to switch to Starburst Galaxy.

The experience with pricing, setup cost, and licensing was straightforward, as I could easily purchase the licensing in the AWS Marketplace. For on-premises, we contacted Starburst for licensing costs, and the setup was quite easy—a one-click setup that allows us to discover the catalogs for querying. The licensing cost was reasonable compared to the value we receive from Starburst Galaxy, making it a good product overall.

I have seen a return on investment; the quick dashboarding allows me to publish dashboards and data products, saving significant time for publishing to different endpoints and translating into a cost savings of around $500,000, which we previously spent on getting reports published due to the reduced turnaround time and our capability to serve many more customers than before.

We did not evaluate other options before choosing Starburst Galaxy; we directly went to it.

I would advise others looking into using Starburst Galaxy to consider it one of the best tools in the market, especially for ingesting large datasets and federating queries across different data sources. Starburst Galaxy is the best engine currently available, and they should try it for themselves to see the difference in query times. My overall review rating is 8 out of 10.

Hybrid Cloud

Amazon Web Services (AWS)