Luna ML Standard
Recommended for pre-production use cases
- 18 tickets per year
- 12x5 support coverage
- 4 hour response time
Experience the power of real-time ML on event data from Kaskada with peace of mind.
Easy to use, high-performance columnar computations over event-based data. Computations like windowed aggregations, over time, for present & historical values.
Operate directly on events in real-time & observe computations on demand or materialized to a table. Execute batch analytic queries over large historical events in seconds.
Easy-to-read, type-safe, declarative queries to compute stateful aggregations, automatic joins, temporal joins, event-based windowing and pipelined operations
Kaskada is a modern, open source compute engine written in Rust and built on Apache Arrow.
Compute event-based features in historical feature computation. Prevent data leakage, or accidental computation of future events that contaminate ML models.
Support engagement from DataStax to help your team develop real-time AI use cases that leverage Kaskada open source.
Kaskada is a unified event processing engine that provides all the power of stateful stream processing in a high-level, declarative query language designed specifically for reasoning about events in bulk and in real time.
Adding an open source component to your ML platform that supports experimentation and production environments can be complex. With Luna ML, customers can deploy Kaskada open source with the benefit of support from the Kaskada experts at DataStax.
All unmodified Kaskada open source components are supported via Luna ML.
No. You don’t need Luna ML to use Kaskada open source; however, Luna ML offers support for your Kaskada deployment from the Kaskada experts at DataStax.
Luna ML offers the peace of mind needed to manage your open source Kaskada deployment in production.
Kaskada’s declarative query language bridges the gap between batch and real time. Match model context in training and production for better performing models—without costly code rewrites.
Depending on your requirements, you can use Cassandra and/or Pulsar as sources and sinks with Kaskada. Luna ML can help you deploy Kaskada with Cassandra and Pulsar.
Apache Cassandra can be an event store (source), feature store (sink) and prediction store (sink). Cassandra offers balanced write and query scalability, low latency, and reliability while supporting large data volumes required by ML at scale.
Pulsar can stream new data into Kaskada as a source and Kaskada can write query results by creating a materialization and writing to Pulsar as a sink in real-time.