KAYTUS Launches Gigawatt-Scale Prefab AI Data Center Solution, Slashes Deployment Time by 60%

June 25, 2026
KAYTUS Launches Gigawatt-Scale Prefab AI Data Center Solution, Slashes Deployment Time by 60%
  • KAYTUS unveiled a gigawatt-scale, fully prefabricated containerized AI data center solution at ISC 2026, engineered for rapid deployment from 3MW up to 1GW.

  • The solution standardizes the full stack into three factory-prefabricated container modules—IT Cube, Power Cube, and Cooling Cube—scaling from 3MW to 1GW and shortening deployment timelines by as much as 60% versus traditional construction.

  • KAYTUS also introduced KSManage Ultra, an intelligent infrastructure management platform for AI Factories, offering full-stack visibility and integrated management of GPUs, racks, PDUs, and CDUs across data centers.

  • KSManage Ultra supports three-level liquid cooling leak detection (node, rack, loop) with automated safety shutdowns, valve closures, node isolation, and alerts for closed-loop remediation in case of leaks.

  • The model is a single-vendor delivery approach covering site survey through ongoing operations, aiming to reduce project management overhead and improve schedule predictability.

  • Integrated management monitors GPUs, CPUs, memory, interconnects, management networks, power shelves, CDUs, and liquid cooling to prevent performance degradation via holistic visibility and correlation analyses.

  • onboarding and operations are accelerated with minute-level capabilities: one-click batch scanning and automatic node addition, topology mapping, batch stress testing, and rack-level automated initialization, reducing onboarding time from about 50 minutes per rack to under 3 minutes.

  • The platform targets three AI-ops challenges: rising rack-scale management complexity, difficulty diagnosing cross-layer performance issues, and inefficiencies from device-by-device onboarding and inconsistent configurations.

  • Key benefits include faster deployment—from roughly 18–24 months down to 6–8 months—reduced on-site risk due to factory integration, single-vendor accountability, higher deployment quality, streamlined lifecycle support, and scalable expansion to 100MW or 1GW.

  • Cooling Cube combines a liquid cooling module and an air-cooled chiller, with redundant equipment, a thermal buffer, and emergency water reserves to maintain cooling during power events or water interruptions.

  • Cooling Cube provides 4,200kW liquid cooling and 3,300kW chilled-water capacity, with redundant pumps and a 72-hour emergency water reserve.

  • The Cooling Cube integrates a high-temperature liquid-cooling source (4,200kW) and a low-temperature air-cooled chiller plant (3,300kW), featuring redundant pumps, water treatment, pressure management, a 10-minute thermal buffer, and a 72-hour emergency water reserve.

Summary based on 5 sources


Get a daily email with more AI stories

More Stories