Announcing FeatureBase 4.3.0

By: Molecula

July 13, 2021 – We are excited to announce the release of FeatureBase 4.3.0! 

This release expands the ways in which FeatureBase can be used by our customers, adding key functionality that improves ease of use, as well as the specific requirements that individual use cases demand.

Enterprise Grade Backup

Backup and Restore

FeatureBase now offers an easy to use built-in command to create backups of feature data. This critical functionality helps customers protect their data and keep their environment operating smoothly while staying compliant with backup and disaster recovery policies.. This backup procedure can happen while FeatureBase is running without the need for downtime or limiting usage.

Use Case Expansion

Process Automation

One superpower of FeatureBase is its ability to analyze massive amounts of unified real-time and historical data in milliseconds. This can have major implications in creating successful process automation within organizations, which is one of the most difficult use cases to deliver on with existing tools. Within the realm of process automation, we are commonly brought in to address network security and operations concerns, including fraud and anomaly detection, proactive maintenance, and scenario forecasting. This release of FeatureBase delivers new functionality to better support process automation, including the ability to leverage an external datastore for deeper, record-by-record analysis, and the ability to delete data in batches.

Our customers often seek further analysis on the insights returned via FeatureBase, requiring the retrieval of events or logs for deeper inspection. FeatureBase excels at filtering through trillions of data points at lightning speed to identify result sets that fulfill a specific criteria. When it comes to extracting individual records associated with query results, key value stores or traditional database systems tend to excel. This release of FeatureBase allows users to get the best of both worlds by leveraging an external data store to return individual records, events, or logs on analytical result sets, when necessary. FeatureBase queries this data store directly and is also able to delete unnecessary data from the data store. This particular release supports Postgres, but expect broadened support in the future!

Speaking of deleting data, use cases that are heavily time-series oriented often require only a certain time frame of data for analysis. While we may try to convince you otherwise (historical data has so much gold in it!), we recognize that easily deleting swaths of data is desired for a number of reasons. In 4.3.0 we deliver the ability to easily delete records via a FeatureBase query.

Integrations

Grafana as a Visualization Tool

Visualize FeatureBase data more efficiently with a new integration with Grafana! Users can now easily download and install our plugin and quickly begin visualizing their FeatureBase data with Grafana. Customers are able to use either PQL or SQL to query their data through this integration.

Figure 1:

FeatureBase Grafana1

[Figure 1 – A Grafana dashboard powered by FeatureBase]

Figure 2:

FeatureBase-Grafana-Visualization[Figure 2 – User experience for building a graph by querying FeatureBase]

Broadened Support for Kafka Ingest

Gain more flexibility when using the Kafka ingester. FeatureBase now supports “static kafka,” which means we can ingest from OSS and other hosted versions of Kafka. This ingester also supports nested JSON arrays and offset-based ID allocation.

Web User Interface

The FeatureBase UI received a couple of major upgrades with the release of 4.3. We now have a no-code query builder that allows users to quickly explore their data via a tabular output (see Figure 3).

Additionally, the FeatureBase UI now surfaces histogram visualizations for Count, TopN, and Groupby queries.

Figure 3:

featurebase-query-builder[Figure 3: Query Builder and histogram visualization in FeatureBase]

With the release of 4.3, we have made major improvements towards enterprise grade backups to create a stable, reliable product for our customers.

We’re excited to see how our customers take advantage of the added functionality and reliability that 4.3 brings! As always, we’d love to hear from you if you have any questions or feedback.

Kudos to the Molecula Product and Engineering teams that led this effort!