Skip to main content

Optimizer

Purpose​

Optimizer — a right-sizing advisor that analyzes each VM's actual resource usage and recommends CPU/memory/disk adjustments, with a one-click apply per VM.

Recommendations are based on observed usage. Auto-Optimize applies that VM's recommendations via the optimize endpoint; skipped changes are reported in the result.

When to use it​

  • Checking whether VMs are over- or under-provisioned before a capacity or cost review
  • Applying a recommended resource change without manually editing a VM's spec
  • Triaging which VMs have high-impact recommendations first
  • After a quiet period of metrics, so recommendations reflect steady state not a spike
  • When Autoscale is too aggressive and you want a one-shot right-size instead

How to get there​

  • Route / id: /resource-optimizer
  • Nav: Monitoring → Optimizer (sidebar, command palette, or desktop nav)

Operate from the console (UX)​

  1. Summary cards: VMs Analyzed, total Recommendations, High Impact, Medium Impact.
  2. If nothing needs changing, the page shows "All VMs are optimally configured."
  3. Otherwise each VM card lists recommendations: resource, current → recommended, reason, impact badge (High / Medium / Low).
  4. Auto-Optimize per VM applies all of that VM's recommendations — button spins, then becomes disabled Applied and reports applied vs skipped counts.
  5. Manual refresh is in the page header.

Typical flow: sort attention to High Impact → review reason → Auto-Optimize one VM → confirm on Virtual Machines / Live Metrics. Respect Quotas before applying large increases.

Empty / fail: Error banner, empty table, or failed toast — confirm you are signed in, zyvor-fabricd is healthy (/readyz on http://127.0.0.1:<port> or https://<host>), and any backend this page needs (FluxVM, storage, network) is reachable. See Admin basics.

Success: Live data loads without error; creates/updates appear in the list or detail view and any confirmation toast clears cleanly.