Databricks Reviews & Product Details

I have been using Databricks platform for business research projects and building ML models for almost a year. It has been a great experience to be able to run analysis and model testing for big data projects in a single platform without switching between SQL server and development environment with Python, R, or Stata. Also, I like the fact that MLflow can track data ingestion for any data shift in realtime for model retraining purposes. Review collected by and hosted on G2.com.

We have had issues using MLflow and feature store on Databricks for ML projects, which slows down the development process. Wish there was better documentation on these tools or more diverse examples to demonstrate different use cases. Also, the test-train split with MLflow does not support time series time interval test-train split for model validation purposes. Review collected by and hosted on G2.com.

It's data processing velocity, storage distribution and compabilty and kind of data Review collected by and hosted on G2.com.

It's overall good. Clusters cost I think it's high Review collected by and hosted on G2.com.

Databricks come with everything in one place, and that is the best thing about the platform. Now I don't have to go to EMR page to create an ERM and then create a notebook. All these can be done with just one click and one page. One gets to jump from notebooks to Query service and to AI in just one click. You get data store views, auto type suggestions based on table and schema context. All these makes the experience very smooth and fast. Review collected by and hosted on G2.com.

Queries get refreshed once in a while, where if some queries are not saved explicitly get deleted.

There is no dark mode, hence have to use a white theme.

Sometimes your queries get pushed to long queues which takes a lot of time and can be little frustrating. Review collected by and hosted on G2.com.

Easy to set up processes and iterate.

Shareability Review collected by and hosted on G2.com.

Not tailored for production integration

Hard to incorporate without being databricks aware, which leads to a vendor lock Review collected by and hosted on G2.com.

The platform is easy to use and collaborate with my colleagues. Deploying to production is simple and can be even easier if you choose the non-self-hosted option. Although Databricks does some of the heavy lifting, it's still open enough to allow teams to use their own flexibility and complex processes without too much configuration. " Review collected by and hosted on G2.com.

Platforms constantly change as they adapt, so staying on top of everything can be difficult. - If you don't have a CICD system in place, once you hit a certain number it starts to get difficult to manage. Review collected by and hosted on G2.com.

It's ease of use and optimisation databricks offers for big data test cases Review collected by and hosted on G2.com.

Some times it's slow and initial learning should be encouraged Review collected by and hosted on G2.com.

Lakehouse makes connecting data from a variety of sources very straightforward and allows our teams to easily merge datasets and annotate data. This makes our processes much more streamlined. Review collected by and hosted on G2.com.

It would be advantageous to have support for various types of developer tools; our use case is internal API development, and it would be beneficial to have the ability to test how schemas can affect downstream API endpoints. Review collected by and hosted on G2.com.

It brings a lot of things together under one area to simplify operations in data engineering and data science. Databricks has enabled me to learn a lot of various topics in data engineering and data science because of their great training modules. The scalability of the Databricks Lakehouse is also very attractive. DLT Pipelines are also a great tool for ETL that I have loved using. Review collected by and hosted on G2.com.

The Databricks Lakehouse is very expensive, although I find it to be largely worth it for our situation because our team and company can utilize all of the tools they offer. Review collected by and hosted on G2.com.

My favorite part abt databricks lakehouse platform would be to have an integrated UI and connecting SQL/DS and ML space. It also has clear instructions on connecting to 3rd party tools. Review collected by and hosted on G2.com.

The help documents sometimes are little hard to navigate and i would need to go thru google search instead. wish there could also be AI modules in the future to answer questions. Review collected by and hosted on G2.com.

It does some compute things nicely and there are some security things in place (but not enough!) Review collected by and hosted on G2.com.

Trying to automate the infrastructure is difficult, and some things are impossible to automate as they are only available on the web console and not through APIs.

Their support is beyond bad. We had to give up working with them as it wasted too much time. Never got a solution in place. Their support is not equipped to debug problems in their own product and they do not give any alternative solutions.

I would look elsewhere for datalake solutions. There are more mature technologies out there without the limitations databricks brings. they also are less expensive and easier to deal with. Review collected by and hosted on G2.com.