Confluent Current 2024: StarTree integrates real-time analytics for observability
StarTree was a partner player at Current, the next-generation of Apache conference, run by data streaming platform company Confluent.
StarTree Cloud, powered by Apache Pinot (an open source, distributed database that provides real-time analytics for low-latency, high-throughput user-facing analytical workloads) is a managed real-time analytics Database-as-a-Service (DBaaS).
Designed for user-facing apps, or backend APIs and microservices, real-time analytics are now a required component powering internal and customer-facing dashboards.
At the Confluent Current conference held in Texas this month, the company showcased new observability capabilities.
Its technology demonstration highlighted how StarTree Cloud can now be used as a Prometheus-compatible time series database to drive real-time Grafana observability dashboards.
Currently, there is a dynamic occurring in the industry as many vendors seek to lock in customers with all-in-one solutions against a countercurrent of open source practitioners looking to select best-in-breed components that comply with open standards, such as Open Telemetry (OTEL).
In the live demonstration, StarTree showed how open telemetry (OTEL) metrics generated by Datadog Vector agents were ingested into Apache Kafka to be consumed by StarTree Cloud.
Disaggregated observability stacks
StarTree Cloud immediately indexed this data to drive real-time observability dashboards in Grafana using the Prometheus query language (PromQL). These capabilities are available today in private preview for StarTree customers and prospects.
“With the world of OTEL, users are exploring new architectures with disaggregated observability stacks. They don’t want a single-vendor lock-in. They want the best components at each layer. The best tooling for their agents. The most flexible ways to visualise their data, or to manage their alerts. For the data storage layer, StarTree provides performance, scalability and affordability for the disaggregated stack. Rather than have redundant, separate data stores built into each observability tool, you can use the same database to manage all of your observability data – all your logs, traces and metrics – as well as your real-time business-oriented data,” said Kishore Gopalakrishna, co-founder and CEO of StarTree.
At Current 2024 StarTree also provided a demonstration of its new Performance Manager (currently in Private Preview).
With Performance Manager, customers can now optimise query performance with a simple guided interface.
When users select “optimise” in the query console the Performance Manager looks at the query structure and resulting query stats and makes a recommendation — whether it is to add an index or bloom filter on a particular column, or add a derived column or materialise certain results in the form of a star-tree Index. What’s more, users can apply the changes directly in the tool and see their performance improve near instantaneously.
The company Tweets at @startreedata