Verdigris + Vertiv | Beyond Environet Thresholds

If you run a Vertiv shop, you already know what Liebert UPS, Environet, and Vertiv's 24/7 services do well. Verdigris adds the validated layer above them. Verdigris EV2 sensors install on circuits and capture 8 kHz waveforms continuously across the full power path; Environet alarms and Liebert telemetry ingest into Verdigris as supplementary context. The result catches equipment degradation 21 days before any threshold-based DCIM system fires an alert. Reliable headroom, not nameplate margin, sets the safe tenant headcount for a hall.

Key takeaways

Vertiv Environet, Liebert UPS, and Vertiv's 24/7 service organization are the current go-forward Vertiv monitoring substrate. Trellis was discontinued by Vertiv (support ended 2023), and no like-for-like successor exists.
Environet is a threshold-based monitoring product. It fires when a measured value crosses a configured limit. Verdigris adds a different measurement floor: 8 kHz waveform analysis that catches degradation before any threshold moves.
Verdigris is vendor-neutral. Liebert UPS, non-Liebert UPS, Vertiv PDU, third-party PDU, and supporting switchgear flow into a single system of record as the substrate.
Stranded capacity recovers when individual-unit physics is validated continuously. Reliable headroom, not nameplate margin, sets safe tenant headcount.
For legacy Trellis sites still on expired support, Verdigris ingests exported telemetry without disturbing the existing deployment.

What Vertiv does well

Vertiv runs critical digital infrastructure for hyperscale and colocation operators. Liebert UPS lines. 24/7 service organizations. Factory-trained technicians. Global field coverage that few competitors can match. The Liebert UPS family has decades of field-proven reliability across colocation, hyperscale, and enterprise sites. Vertiv's service depth (phone dispatch on critical alarms, predictable response times, equipment-specific engineering) is what colocation operators are buying when they standardize on Vertiv.

On the software side, Vertiv's current go-forward monitoring is Environet (acquired in 2018 and positioned for power, thermal, and environmental telemetry). Environet Alert speaks SNMP, Modbus TCP/IP, BACnet/IP, and serial via IP converters. It is sold per-device-license. The scope is narrower than the previous Trellis DCIM platform. Vertiv discontinued Trellis and ended support for existing contracts in 2023. Existing Trellis installations may still be in the field on expired support, but Trellis is not Vertiv's go-forward DCIM surface.

For an operator running a Vertiv-standardized estate today, this stack (Environet for threshold monitoring, Liebert UPS hardware, and Vertiv 24/7 services) is the natural operations substrate. The question is what intelligence sits above it, what looks at the entire power path, and what measurement floor catches the degradation that threshold systems are not built to see.

Why not extend Environet + Liebert telemetry to higher frequency?

Even if Environet polled at higher frequency, it would still be threshold-based — the wrong analytics regime for waveform-physics anomaly detection. Frequency increase doesn't change the analytics architecture. The ceiling is in what is captured, not how often it is sampled.

Environet polls field protocols (SNMP, Modbus TCP/IP, BACnet/IP) at threshold-monitoring cadence. Liebert UPS hardware exposes its own SNMP and Modbus telemetry at device polling rates. Neither retains continuous waveform; both report state when configured limits cross. A faster threshold poll still reports state, not waveform.

The validated layer is the second gap. Pretrained ML for rectifier degradation, capacitor ESR drift, and harmonic ratio shifts is not a feature inside Environet or the Liebert UPS firmware. That analytics layer does not exist on Vertiv's go-forward platform.

Verdigris EV2 sensors clamp on input circuits and capture 8 kHz continuously, independent of any vendor sensor's native capability. The intelligence flows back to your Environet alarms and Vertiv 24/7 services workflows via standard protocols.

How does Verdigris integrate with Vertiv Environet?

Environet Alert and similar threshold systems work by setting a tripwire. When a measured value (branch current, PDU temperature, UPS battery voltage, environmental humidity) crosses a configured limit, an alert fires and a service ticket is created. This is the standard architecture for threshold-based monitoring, grounded in setpoints rather than waveform physics. It works well for failure modes that announce themselves with a clean threshold crossing: battery voltage drops below a setpoint, capacitor temperature exceeds a band, a fan slows below RPM minimum.

"Threshold-based alarms are good at telling you what already broke. They are not built to tell you what is degrading inside the band."

The architectural limit shows up when degradation evolves at the threshold but never crosses it. In a deployment of 800+ rectifiers documented at T-Mobile, Verdigris detected active equipment degradation in approximately 4 percent (about 32 units) of the fleet. Zero standard alarms had fired on those units. The degradation lived in 5th and 7th harmonic ratios and in capacitor ESR drift. It did not live in any threshold value Environet or a Liebert-native alarm would have been configured to monitor. The threshold was the wrong measurement to watch.

"Verdigris has been instrumental in refining our capital planning processes, enabling us to make more informed and strategic investment decisions across our facilities."

John Coster, Sr. Manager, Innovation, Planning and Strategy at T-Mobile

This is not a critique of Vertiv's products. It is a critique of threshold-based architectures generally. The Uptime Institute has documented how alarm-driven monitoring optimizes for failure detection rather than failure prediction. IEEE 519 names total harmonic distortion (THD) as a leading indicator of power quality equipment health long before any standard threshold breach. The EPRI power quality body of work tracks the same pattern across utility and critical-load assets. IEC 61557-12 defines the instrument-class accuracy required for power quality measurement. Catching pre-threshold degradation requires a different measurement floor: continuous waveform capture at 8 kHz, analyzed by ML trained on the harmonic signatures of failing equipment.

Vendor-neutral across the full power path

Environet monitors Vertiv devices. Liebert telemetry comes from Liebert UPS systems via SNMP and Modbus. This is by design and it is the right scope for a hardware vendor's monitoring platform. But most data centers do not run a single-vendor power chain.

A typical NeoCloud or enterprise hall runs Vertiv Liebert UPS, Schneider or Eaton switchgear, ABB or Siemens breakers, Honeywell BMS overlays, and mixed PDU vendors at the rack. The power path crosses three or four monitoring islands. Each speaks its own protocol. No shared canonical view of the chain exists. When a transient propagates from a rectifier through the busway and trips a downstream breaker, no single vendor's monitoring sees the whole story.

Verdigris EV2 sensors clamp on any circuit regardless of brand and capture 8 kHz waveform data continuously across the full chain. The 8 kHz lives on Verdigris's own physical sensors, not inferred from Vertiv telemetry. Environet alarms and Liebert UPS telemetry ingest into Verdigris alongside that native waveform stream as supplementary context. As the substrate, Verdigris becomes the system of record. Environet and Liebert continue running as device-level inputs; Verdigris correlates them with the rest of the power path.

How Verdigris adds value to your Vertiv estate

The integration is additive. Environet, Liebert telemetry, and Vertiv 24/7 services continue running exactly as they do today. Verdigris EV2 sensors install non-invasively on the input circuits to capture 8 kHz continuous waveforms. Verdigris ingests Environet alarms and Liebert telemetry as supplementary context, then provides validated operating limits and early warning across the entire fleet.

Data Source	Vertiv System	Direction	Verdigris Role
8 kHz waveform on circuits	Verdigris EV2 sensors	Native capture	Continuous, vendor-neutral measurement floor
UPS telemetry	Liebert UPS via SNMP / Modbus	Inbound to Verdigris	Cross-checked against EV2 8 kHz waveform on the input circuit
Environmental and PDU alarms	Environet Alert (threshold-based)	Inbound to Verdigris	Filtered and validated against Verdigris early-warning signals
Critical alerts and dispatch	Vertiv 24/7 service organization	Outbound from Verdigris	Pre-threshold tickets fire 21 days earlier on validated physics
Legacy Trellis telemetry (where present)	Trellis (legacy / unsupported)	Inbound to Verdigris	Ingested via exported feeds; no dependency on continued Trellis support

Integration with the operator's existing surfaces is outbound. Verdigris intelligence flows out via four channels:

MQTT broker (Verdigris-hosted) for real-time pub/sub
REST API for programmatic access to electrical state and alerts
Modbus for production-validated OT integration
BACnet emulation. Verdigris sensors appear as native BACnet devices on the operator's existing network. Zero integration lift: the operator's BMS sees Verdigris as just another BACnet device, no BMS config change required

The operations team keeps using whatever surface they already use. Environet keeps firing on threshold breaches. Vertiv 24/7 keeps its critical-alarm response posture. The Verdigris intelligence layer sits underneath, validating physics that lives between threshold setpoints.

The intelligence Verdigris adds

21-day advance warning. Verdigris flagged degrading T-Mobile rectifiers an average of 21 days before any threshold-based system would have fired an alarm. That window converts unplanned outage exposure into scheduled maintenance. It also gives the Vertiv 24/7 services team a planned dispatch window instead of a reactive critical-alarm response.

Individual-unit physics validation. Each unit is evaluated against its own harmonic and waveform expectations. A unit whose telemetry sits comfortably inside Environet's threshold band can still be flagged when its electrical signature drifts from physics-expected behavior. The evaluation is independent of the threshold setpoint.

Vendor-neutral coverage. Liebert UPS, non-Liebert UPS, Vertiv PDU, third-party PDU, and supporting infrastructure all flow into a single system of record as the substrate. The detection logic does not care about the brand on the cabinet.

Validated operating limits. Verdigris computes capacity headroom that threshold-based monitoring does not produce: how much additional load a given Liebert UPS or PDU can safely carry, grounded in the physics of its current condition rather than its nameplate or its threshold band. This is the validated layer that turns monitoring into decisions. It also turns nameplate-derated capacity into actual, defensible tenant headroom.

How it deploys in a Vertiv shop

A colocation operator runs Liebert UPS systems across two halls, with Environet for environmental and PDU monitoring and Vertiv 24/7 services on call for critical alarms. The operator adds Verdigris EV2 sensors on the input circuits without taking any equipment offline. Installation completes in days. Environet keeps polling Vertiv devices. Liebert UPS keeps reporting SNMP and Modbus telemetry. The operations team keeps using the surfaces they already use.

Verdigris turns on BACnet emulation, and the EV2 sensors appear on the operator's existing BMS as native BACnet devices. No BMS reconfiguration, no new dashboard for the team to learn. Verdigris MQTT and REST feeds open in parallel for any automated workflows the operator wants to wire in later.

Environet continues firing on threshold breaches. Vertiv 24/7 keeps its critical-alarm response posture. Verdigris waveform analysis flags four specific rectifiers degrading via harmonic-content shifts, all with telemetry sitting well inside Environet's configured thresholds. Service is scheduled 21 days before any threshold would have moved. The Vertiv field services team executes the work on planned maintenance windows. No tenant SLA exposure, no after-hours dispatch, no surprise. The operator recovers stranded capacity that nameplate-margin assumptions had been holding in reserve.

For operators still running Trellis on legacy or expired support, Verdigris can ingest exported telemetry without disturbing the existing deployment. The validated layer does not depend on Trellis remaining a supported product through Vertiv's sunset of the platform.