Helix is the first instant, responsive data engine that creates a dual backbone of modern business intelligence and interactive data science.
Mode automatically streams query results into Helix, a high-performance in-memory data engine, so that you can visually explore up to 10GB — faster and with less database load.
Every Mode query result is immediately streamed into Helix.
Visually explore tens of millions of rows.
Stakeholders can explore results with drag-and-drop visualization tools.
Helix is awesome. Changed my life. I was able to visualize more than 4.6M rows of data with no performance issues for the first time.
Because stakeholders can extend any analysis to answer their own questions, data scientists no longer need to predict what they’ll be asked next.
A high-performance, in-memory database designed for filtering, aggregating, and manipulating query results with sub-second latency. Every Mode query result is automatically streamed into this engine as soon as the query is completed. Once loaded, this data can be visualized in charts or tables. Helix can currently visualize 2,000 times more data than previous limits.
No. Helix is an in-memory database where we load query results to enable further analysis on them. It's an on-demand data extract, not a data warehouse.
No. We don't clone your tables and store them into our own data warehouse.
Helix works with any database that connects to Mode.
Here is a list of the databases we support: https://mode.com/data-sources/
Today, data in Helix is immutable. If you want to refresh the data, then re-run the query. The new query results will be loaded into Helix the same way the prior query results were.
When the customer hasn’t done any analysis on some data for a while, we remove it from the in-memory database to make room for other data sets.
Helix is built exclusively for interactive queries.
No. The mechanics work equally as well with any underlying database. Mode doesn't sync your database, instead it uses your SQL query as an instant extraction script. This ends up being a lot faster than just connecting your BI to your full-scale data warehouse.