Skip to main content

Product Roadmap

We are focused on building an integrated observability tool which can be a superior alternative to current SaaS products like DataDog

More longer term, we plan to ship the following items:

Different retention period for different sources of logs data​

Different types of data have different retention requirements. Some applications need to store logs for 1 week, some for 1 month and some for 1 year. We will be building a solution to enable this.

Multi-Tenancy​

One company should be able to onboard different teams which have different datasets. Multi-tenancy enables users to manage multiple datasets in the same database without doing multiple installations of SigNoz.

Cost Control​

  • Monitoring and limiting team-wise quota
  • Check on cardinality explosion in metrics and suggestions of deletion of unused attributes in timeseries
  • Tail based sampling on latency, error or any attribute in spans and logs

Tail based sampling​

The decision of which spans to sample is crucial in distributed tracing. "Tail-based sampling" means that trace-retention decisions are done at the tail end of processing after all the spans in a trace have arrived.

We plan to focus on this to ensure that only 'interesting' traces are ingested. This will ensure that we don't miss any interesting traces, even though the actual number of traces ingested/retained is much lower.

Anomaly detection framework​

A framework to provide dynamic thresholding capabilities to enable better Signal to Noise in alerts. One of the projects we are closely following in this space is LinkedIn's Third Eye If you know of any other good frameworks, please share in Github Discussions

We believe in taking feedback from our community. Feel free to jump to our Github Discussions if you have any idea or feature we should build first. We are all ears 👂👂