Skip to content

Capacity Trends

gitsync edited this page Jul 14, 2026 · 4 revisions

The Capacity Trend page answers "how fast am I filling up, and where?" with long-term charts, a folder-size TreeMap, and a live Topology map of your whole storage/compute/network landscape.

Capacity Trending

Capacity Trend

Unlike the 24-hour dashboard, capacity data is stored for up to 5 years, letting you spot growth trends and plan upgrades.

  • Pool/dataset selector with quick "pool + all datasets" buttons
  • Time range: Day / Week / Month / Year / Since Data
  • Stacked area chart with a dashed ceiling line showing total usable capacity — when the stack approaches the line, it's time to expand
  • More than 6 datasets selected? The smallest are grouped into an "Other" curve to stay readable
Pattern Likely meaning
Steady upward slope Normal growth
Sudden step up Large import or snapshot accumulation
Flat line Dataset at its quota
Downward step Snapshots deleted or data removed

Data retention: 5-minute samples for a week, 30-minute for a month, daily for 5 years — self-capped around 10 MB on disk (config/capacity.rrd.json).

Folders TreeMap

The Folders TreeMap tab renders folder sizes as a treemap — the fastest way to see which directories consume your space. Scans can be run on demand or on a schedule per dataset.

Topology

Topology map

The Topology tab draws a live map of everything the portal knows, on two switchable layers:

  • Storage & Compute Layer — remotes (InterLink peers), SMB/NFS shares, VMs & containers, datasets/ZVols, pools, and physical disks — connected by their real relationships, with live I/O rates on the links
  • Networking Layer — bridges, NICs, VLANs, and instance network attachments (see Virtual Networking)

Features: zoom controls + fullscreen, draggable nodes with remembered positions, hover cards with live metrics and per-disk iostat, and click-through navigation to the underlying object. Active SMB/NFS client counts appear on share nodes.

Related pages

Clone this wiki locally