Blog

evitaDB 2026.2 brings fundamental changes to the write side of the database. New granular, persistent data structures dramatically cut the amount of data that has to be copied, relieve the pressure on the garbage collector and — together with storing only the parts of an index that actually changed — make datasets in the millions of records comfortable to work with. Beyond higher write throughput, the release widens the control you have over transaction conflicts, fixes the atomicity of entity-level operations and adds histograms over reference attributes and over range values, aimed at advanced filtering in e-commerce applications.

Agentic programming is fascinating, but it has one fundamental weakness — the agent can't see what's actually happening inside a running application. We built an MCP server that gives AI agents full access to the Java debugger via the JDWP protocol. In this article, we'll look at how it works, how you can try it yourself, and why it helps us in developing evitaDB.

We are preparing a new version of our website, a significant part of which is user and developer documentation. Maintaining documentation in more than one language is beyond our capacity, so we chose English, but the ambition to provide documentation in developers' native languages never left us. In this article, we offer a look at the mechanism we want to use to achieve this.

The latest version of evitaLab introduces a new feature for browsing and searching the Write-Ahead Log (WAL), enabling detailed exploration of all database changes. In this article, you'll learn how to use this tool effectively and what it's designed for.

The end of the year is a perfect moment to look back at how far evitaDB has come. The key theme of this year was stability and adapting to real-world production needs, which we see in greater numbers every month.

Change Data Capture (CDC) is a pattern for detecting and streaming inserts, updates, and deletes from a source database in near real-time. It enables low-latency data replication, event-driven workflows, and keeps services in sync without heavy batch jobs. If you're wondering how CDC is implemented in evitaDB and how we ensure its reliability, read on.

Data writes in evitaDB are specific, as it treats data on disk as immutable. All changes are appended incrementally to the end of files. This approach has its advantages and disadvantages. One drawback is the need for data compaction, as files gradually fill with outdated records. Compaction happens synchronously during transaction processing once certain conditions indicating an excessive amount of outdated data are met. This can impact transaction commit latency. This article discusses how clients can respond to this scenario.

Our web-based GUI client evitaLab has proven to be an irreplaceable tool when working with evitaDB. The one major thing that was missing in our daily usage was an easy way to connect to different evitaDB servers. Therefore, from the very beginning we wanted a desktop version that could connect to any evitaDB server from a single place.

Working with "one-to-many" references is very common in data models and presents a variety of problems that are not immediately obvious. If the database itself does not provide enough tools and expressive means to solve them, the application developer is forced to address these issues themselves, which usually leads to a significant increase in workload and performance problems. In version 2025.2, evitaDB introduces new ways to work with these types of references, helping to address common scenarios in e-commerce applications.

Imagine you could record all the traffic on your database, analyze it and replay it. How much easier would your life be? It's surely worth some effort to set up proper tooling to make this possible. But, what if I tell you, that all this magic can be done with just turning on a single switch?

Native soft-delete support in evitaDB simplifies application data models, enhances performance, and reduces complexity by handling archived entities directly within the database.

We're excited to introduce a powerful new feature in evitaDB — the priceDiscount ordering constraint. This feature allows you to sort your product listings based on the discount amount, helping you highlight the best deals to your customers. In this post, we'll explore how to use this feature effectively, with real-world examples, practical query snippets, and insights into different pricing strategies, including handling products with variants and product sets.

evitaDB started its journey with the Undertow server from JBoss. It was chosen because of its performance and ease of use. However, due to the gRPC API, we still had to run a separate Netty server under the hood. This was far from ideal, and we hoped that with Undertow 3.0, which promised to move to Netty, we could unify all APIs under one server. Unfortunately, the development of Undertow 3.0 has not progressed in years and we had to look for alternatives.

evitaLab is still being actively developed alongside evitaDB and there are many new features that are worth mentioning. There are features for learning evitaDB as well as for debugging data.

I have always used the semantic versioning (SemVer) scheme for my projects. However, after discussing with my friend Lukáš Hornych and conducting further research with the team, we have decided to change the versioning scheme for evitaDB. The new scheme will be a Calendar Versioning. If you are interested in following our debate, read on.

A lot has happened this year - and a lot is being prepared for next year. For those who want to follow the development of the project, we have prepared a summary of the most important events of the year 2023 and a brief overview of the plans for next year.

It's always a bit painful to troubleshoot a problem with your production data inside a database if you don't have a handy tool to look inside easily, and instead, you have to write different kinds of queries using different tools to get the information that you need. We feel the same way. That's why we've created a hopefully handy GUI client for evitaDB called evitaLab, so you have a buddy when you're either just playing with evitaDB or you're already debugging production data.

This article is a continuation of a series of articles about the documentation portal of our database. This time we will look at it from a different point of view - from the perspective of the developer who creates the documentation, and whose interest is that the documentation written is correct and does not become outdated over time.

The documentation on evitaDB site is getting bigger and bigger. The more examples we add, the more we're afraid they will become obsolete or broken. How do we tame this beast of hundreds of examples?

A fast test suite is a critical aspect that motivates developers to write more tests and run the suite frequently. The ideal test suite should finish in a matter of seconds or low units of minutes.