Nexus World · Data Center Monitoring (DCIM)
Data Center Infrastructure Management

Every watt, every rack, in real time

Nexus·DCIM monitors your entire data center — the full power chain, cooling, environment, assets and capacity — from one supervisory platform, so you catch problems in seconds, not post-mortems.

NEXUS·DCIM — NOC OVERVIEW LIVE 1.32 1.0 2.0 PUE · FLEET AVG IT LOAD 842 kW ▲ 12 kW · 1h COOLING LOAD 316 kW ▼ 4 kW · 1h UPS LOAD · N+1 78% Batteries 100% · OK DATA HALL A — RACK INLET °C < 24 °C normal 24–27 °C warning > 27 °C critical TOTAL FACILITY POWER — 24 H Rack A-14 inlet 27.9 °C — warning raised 13:02 UPS-2 returned to normal · 12:41 CRAC-03 filter maintenance due in 5 days

99.982%

Uptime maintained

1.32

Fleet-average PUE

10k+

Monitored points

200+

Live sites worldwide

End-to-End Visibility

The full power chain, instrumented

From the utility intake to the last rack outlet — every stage is metered, trended and alarmed, so a failure anywhere in the chain is located in seconds.

Utility11 kV intake
→
GEN / ATSauto changeover
→
UPSN+1 · monitored
→
PDUbranch metering
→
Buswaytap-off meters
→
Rack PDUper-outlet kW
→
IT Loadserver level
DCIM Modules

Everything a data center needs watching

Modular by design — start with power and environment, grow into full asset and capacity management on the same platform.

Power Monitoring

Utility, generator, UPS, PDU, busway and rack PDU — metered to the outlet. Live kW, kWh, power factor and harmonics with capacity headroom per rack and per breaker.

Per-outlet metering Breaker capacity PUE / DCiE
breaker limit

Environmental Monitoring

Temperature and humidity at every rack inlet, hot/cold aisle mapping, water leak ropes under raised floors, airflow and smoke sensing.

Inlet °C / %RH Leak detection

Asset & Capacity

Every rack, server and cable documented with live U-space, power and port capacity — plus what-if placement planning for new loads.

Rack elevations What-if planning

Cooling & CRAC Control

CRAC/CRAH status, supply and return temps, chilled-water plants and containment — tuned continuously to shave PUE without risking hotspots.

CRAC sequencing PUE optimization

Battery Monitoring

Cell-level voltage, internal resistance and temperature for every UPS string — failing cells flagged months before they threaten the load.

Cell-level SOH Predictive alerts

Alarms, Incidents & Reporting

One alarm console with escalation to SMS, email and phone. Every event time-stamped for root-cause analysis, plus scheduled SLA, energy and PUE reports for management and tenant billing.

Escalation chains SLA reports Tenant billing
CRITICAL · PDU-2B breaker trip WARNING · Rack A-14 inlet 27.9 °C CLEARED · UPS-2 output normal
NEXUS·DCIM — ASSET VIEW · HALL A RACK A-14 SW-CORE-01 PATCH-A SRV-A01 · 2U SRV-A02 · 2U SRV-A03 · 2U SRV-A04 · 2U SRV-A05 · 2U STORAGE-A · 3U UPS-FEED A+B 38 / 42U · 6.2 kW RACK A-15 SW-TOR-15 SRV-B01 · 2U SRV-B02 · 2U SRV-B03 · 2U SRV-B04 · 2U SRV-B05 · 2U RESERVED what-if: GPU node 4U 29 / 42U · 4.1 kW 24.1 °C 21.9 °C HALL A · CAPACITY U-SPACE 1,214 / 1,680U POWER 842 / 1,200 kW COOLING 68% headroom NETWORK PORTS 1,872 / 2,304 FLOOR WEIGHT 61% used WHAT-IF RESULT GPU node fits A-15 power OK · cooling OK 4U reserved · ports 8/8
Asset Intelligence

Know exactly what's in every rack

Live rack elevations replace stale spreadsheets. Every server, switch and cable is tracked with its real power draw and port usage — and before new hardware arrives, what-if planning finds the safest slot for it.

  • U-space, power, cooling, ports and floor-weight capacity tracked per rack and per hall
  • What-if placement checks power, cooling and network before you rack a single server
  • Full audit trail — every move, add and change logged with who, what and when
SNMP Modbus BACnet/IP IPMI · Redfish MQTT REST API
Server racks in a data center
Get Started

See your data center like never before

Book a free assessment — we'll map your power chain, place the sensors that matter, and stand up a live DCIM dashboard for your facility.