--- title: "Device Health & Rightsizing Dashboard Overview" slug: "device-health-rightsizing-dashboard-overview" updated: 2026-04-07T13:02:40Z published: 2026-04-07T13:02:40Z canonical: "support.controlup.com/device-health-rightsizing-dashboard-overview" --- > ## Documentation Index > Fetch the complete documentation index at: https://support.controlup.com/llms.txt > Use this file to discover all available pages before exploring further. # Device Health & Rightsizing Dashboard Overview This article introduces the Device Health & Rightsizing Dashboard in ControlUp ONE and explains how Device Health Scores and rightsizing recommendations are determined. ## What is the Device Health & Rightsizing Dashboard? The Device Health & Rightsizing Dashboard provides a unified view of device health across your fleet. It aggregates performance, stability, and hardware lifecycle data into actionable recommendations for rightsizing, hardware replacement, and maintenance prioritization. A health score (0.0–10.0) is used to stack rank devices and identify those requiring immediate attention. ![Health and sizing dashboard UI](https://cdn.document360.io/098100b7-b9da-4bea-b4b9-017140ab863e/Images/Documentation/DeviceHealthDashboard(1).png) ## Where to Find the Dashboard 1. In ControlUp ONE, go to **Dashboards** in the left navigation. 2. Open the **Dashboard Gallery** tab. 3. Select the **Device Health & Rightsizing Dashboard** dashboard. Required license The dashboard is available if you have a ControlUp for Desktops Advanced or ControlUp ONE platform license. ## What is a Device Health Score? The Device Health Score is a single number from **0.0 to 10.0** that reflects the overall condition and suitability of a device for its workload. | **Score range** | **Label** | **Meaning** | | --- | --- | --- | | **9.0 – 10.0** | Excellent | Device is performing well; routine monitoring is sufficient. | | **7.5 – 8.9** | Good | Device is working well with minor issues. | | **6.0 – 7.4** | Fair | Some problems should be addressed; plan optimization or upgrades. | | **4.0 – 5.9** | Poor | Significant problems affecting performance; consider upgrades or replacement. | | **0.0 – 3.9** | Critical | Serious problems; immediate investigation and likely replacement needed. | Scores are calculated using a **Penalty-First** model: devices start at 10.0 and points are deducted based on issues detected. The score is based on a **30-day trailing window** of active user session data to smooth daily fluctuations and improve confidence. **Important:** If a device’s operating system is **Unsupported** (End of Life), the maximum health score is capped at **7.0** regardless of other factors, to reflect security risk. ## What Makes Up the Device Health Score? The score is derived from five weighted pillars. Data is evaluated **only during active user sessions** so recommendations reflect real usage. | **Pillar** | **Weight** | **What it measures** | | --- | --- | --- | | **Rightsizing** | 30% | Whether CPU, RAM, Disk, and GPU usage during active sessions indicates undersizing, appropriate sizing, or oversizing. | | **Stability** | 20% | Reliability: crash/freeze frequency (Windows: MTBF; macOS/Linux: sustained resource pressure as a proxy). | | **Battery Health** | 25% | Battery capacity vs. design; penalties apply when health is below 80%. | | **OS Lifecycle** | 15% | Whether the operating system is still supported; Unsupported status applies a full penalty and caps the score at 7.0. | | **Storage Space** | 10% | Free disk space; penalties apply when free space is below 15%. | ### The Five Pillars in Detail #### Rightsizing (30%) Rightsizing evaluates **breach** (resource stress) and **idle** (underuse) during active sessions: - **Breach (Undersized risk):** CPU or GPU > 80% utilization; RAM available < 10%; Disk active time > 60%. - **Idle (Oversized potential):** CPU or GPU below 40% and 10% respectively; RAM available > 60%; Disk active time < 10%. The pillar uses a quadratic penalty so that penalties increase more sharply as devices approach the 15% breach threshold. A **Risk-First** rule applies: a device with any resource stress is never labeled **Oversized**, even if other resources are idle. #### Stability (20%) - **Windows:** Mean Time Between Failures (MTBF) for app crashes and hangs; BSOD/system crash incurs a full penalty. - **macOS/Linux:** Sustained CPU and RAM breach is used as a stability proxy when native crash data is not available. #### Battery Health (25%) Penalties increase as battery health drops below 80% of design capacity. Devices with no battery (e.g. desktops) receive no penalty for this pillar. #### OS Lifecycle (15%) Operating system support status is evaluated against a lifecycle database: - **Supported:** No penalty. - **Expiring Soon (within 90 days):** Informational; no penalty. - **Unsupported (EOL):** Full penalty; **total score capped at 7.0**. #### Storage Space (10%) Penalties apply when free disk space falls below 15% of total capacity, with severity increasing as free space decreases. ## Rightsizing Recommendations In addition to the health score, the dashboard provides a **Sizing Recommendation**: | **Recommendation** | **Meaning** | | --- | --- | | **Undersized** | At least one resource (CPU, RAM, Disk, GPU) breached its threshold more than 15% of active session time. | | **Appropriate** | Resources generally meet demand without consistent stress or idle. | | **Oversized** | No breaches and all four resources were in the idle band for more than 90% of active session time (requires sufficient data). | | **Insufficient data** | Fewer than 100 active session samples; no recommendation is given. | ## Confidence Levels Recommendations and scores are tagged with a **Confidence** level based on how much active session data was collected in the analysis window: | **Level** | **Active session samples** | **Impact** | | --- | --- | --- | | **High** | > 1,000 | Full scoring and recommendation. | | **Medium** | 100 – 1,000 | Full scoring and recommendation. | | **Low** | < 100 | Sizing recommendation is shown as **Insufficient data**. | Samples are typically 1-minute intervals during active sessions; 100 samples is roughly 1.6 hours of active use. ## Top Issue The dashboard identifies a **Top Issue**—the single biggest factor affecting a device’s health—to help prioritize actions. Typical values include **OS EOL**, **OS Near EOL**, **Battery Health**, **Stability/Crash Events**, **Low Disk Space**, **High RAM Pressure**, **High CPU Pressure**, **Resource Constraints**, or **None**. ## Data Sources and Storage The dashboard compiles data from various indices. The raw data is processed and saved in the `device_health_ddc` index. Data for the dashboard is updated once per week.