Currently, hourly and nightly tally operations are triggered by OpenShift cron jobs. Because the nightly tally will soon be based on SWatch Event records (like the hourly tally), we should investigate whether moving both operations to a pure event driven architecture makes sense now, instead of the planned phase to continue with the nightly job.

Moving to an "event driven tally" would mean that host and usage updates would be applied as Swatch event messages are received/processed, with minimal delay, instead of waiting for the scheduled jobs to run both hourly and nightly.

Acceptance Criteria

• The design should keep testability in mind at each stage.
• Recommend pairing on this task - 2 devs + 1 QE.
• A design document is produced:
  • Details how a single stream of event messages can be processed to update hosts and snapshots in 'real time' for both nightly and hourly tally.
  • Should consider prioritizing hourly events since HBI based tally can have a delay.
  • Defines how each stage of implementation can and should be tested.
  • Should keep both the hourly and nightly tally in mind.
  • Should document any side effects that might impact the UI (eventual consistency).
• Design is presented to the team.