[APM] Surface Slow transactions and errors correlation analysis

[nehaduggal]

Summary of the problem (If there are multiple problems or use cases, prioritize them)
As a user I want to quickly view interesting data about my slow transactions or errors.

User stories
As a DevOps/SRE engineer at my company, responsible for keeping our production systems running I’d like to understand why certain traces are slow. I'd like to see if there is metadata/tags with these traces that is different from the normal traces which could point me to the root cause of these slow transaction.

Examples of this could be particular hardware (host / pod) that is failing and therefore resulting in slow transactions. Another reason could be a set of users (based on ip address or region) who are physically far away from the nearest datacenter and is therefore experiencing increased latency.

[formgeist]

Design update - Dec 22, 2020

I have been meaning to share a design update for a while on this project since it's been developing quite fast over the past months or so. We have been polishing a POC implementation of the significant terms which is available in 7.11 (feature-flagged). Additionally, we have been redesigning basically all of the Transactions views to align them with the new Service overview, and add new capabilities beyond just the significant terms feature, but also new distribution charts and a dedicated trace view.

The placeholder implementation issues have already been created due to planning and I will begin to fill these out with the details for significant terms and the redesigned transactions list.

🎬 Watch the video
🕹 Figma prototype

More screens

Transactions list

Transaction group detail

Trace view

Significant terms flyouts

cc @nehaduggal @alex-fedotyev @sqren

[formgeist]

Closing in favor of implementation issues elastic/kibana#86477, elastic/kibana#86783, and elastic/kibana#86478