Hyperscale
Data Centers
Many of the world’s most prominent organizations rely on hyperscale data centers to support their massive data storage needs and international customer bases.
These business-critical facilities must maintain constant uptime and availability, and because of their huge scale and high-density infrastructure, traditional monitoring solutions are not feasible. Broad visibility, operational efficiency, and process automation are especially important for hyperscalers.
Primary Cooling Units
Chillers and air handlers are exposed to unexpected conditions. To ensure continuous operation, Parameter installs sensors at strategic fail points to detect leaks and other mechanical issues.
UPS & Battery Banks
UPS systems support the critical power needs of your facility. Parameter provides continuous monitoring on your system’s battery banks to detect failures long before they occur.
Reduce system vulnerability by 90% and prevent all critical battery string failures to ensure uptime reliability.
Gen Sets
Maintaining healthy generator batteries is crucial to ensure your facility can survive a utility power failure.
UPS systems are critical to protect against the first moments of vulnerability, but most only last for a few minutes. A single cell failure can compromise an entire string of generator starter batteries, leading to prolonged downtime.
Raised Flooring
We offer many solutions for facilities with raised flooring, from under-floor leak detection cable and airflow sensors, to our own proprietary raised flooring system. Our raised flooring solutions prioritize ease of service and installation while achieving better turbulent flow and air distribution than our competitors and resulting in 5-15 degrees lower server temperature.
Liquid Cooling
When using liquid cooling in a high-density server application, failure safeguarding is essential. Parameter’s RLE Leak Detection has been leading the industry for decades, and that innovation is why we are the go-to solution for hyperscale liquid cooling leak detection.
Distributed Architecture
Our high-density monitoring solutions are structured in a distributed architecture with a dedicated controller in each server rack that can monitor critical parameters, and send commands for valve or server shutoff to prevent hardware or asset loss in case of an emergency.
All sensors and controllers are consolidated to integrate seamlessly with your facility’s native BMS system.
Air Cooling
Parameter offers extensive support and expertise for air-cooled systems. Our wireless temperature and humidity probes are easy to install, and integrate easily with your BMS via our WiNG Manager, consolidating all data and presenting alerts on a map-view interface for easy threat location.
Hot / Cold Aisle
Maintaining proper hot and cold aisle temperatures is key to optimizing performance as well as preventing failures.
Parameter’s environmental monitoring solutions provide easily deployable and scalable temp / humidity readings.
BMS Integration
Parameter systems are purpose-built to minimize the time from hazard detection to resolution.
All our solutions integrate seamlessly with your native BMS regardless of protocol, and sensor data is securely hosted on a local integrated web server to make alerts immediate and actionable.
Switchgears
Switchgear Batteries are often overlooked as failure points in critical systems. These batteries can remain on standby for long periods of time and only show signs of failure when they are needed during an outage risk event.
Continual monitoring of all critical battery strings is vital to ensure facility uptime.
Parameter Applications for Hyperscale Data Centers
Click inside to view how Parameter systems integrate with critical infrastructure to provide the highest quality monitoring solution for hyperscale data centers.
Monitoring Considerations
Multisite Scalability
Hyperscale data centers need to process massive volumes of data for international end-users. Because of this, they typically work across many sites, sometimes up to hundreds globally. They need scalable and flexible solutions that can accommodate multisite architecture and quick development.
Uptime SLA Compliance
Uptime Service Level Agreements (SLAs) require hyperscalers to commit to a certain minimum percentage of uptime for their users. Unexpected power outages compromise their ability to meet these requirements, especially when UPS systems fail, but an effective monitoring system enhances SLA compliance by alerting to potential UPS failures before they happen.
Secure Remote Visibility
With multiple sites and extremely equipment-dense server rooms, it is not practical for hyperscalers to oversee operations in person at all times. A monitoring system that enables secure remote visibility empowers hyperscalers to track important metrics across their entire infrastructure from anywhere.
Automation & Integration
Hyperscalers require monitoring solutions that integrate seamlessly with native BMS through centralized data collection and firmware that can be adjusted easily without infrastructural overhauls.
Failure Points
Liquid Cooling
The high-density AI/ML clusters and GPUs housed in hyperscale data centers are beginning to require exceptionally efficient in-server liquid cooling methods to prevent overheating. However, the intricate networks of flow valves involved in these methods create more opportunities for coolant leaks near sensitive equipment.
Large-Scale UPS Systems
Hyperscalers maintain constant uptime with large-scale Uninterruptible Power Supply (UPS) systems that support critical functions when primary power sources aren’t available. The large volume of loaded battery cells contained in these systems increases the risk of unexpected failures. To prevent microfailures, battery monitoring systems must be comprehensive enough to cover all cells across massive UPS systems.
CRAC/CRAH Units
Computer Room Air Conditioning (CRAC) and Computer Room Air Handler (CRAH) units regulate ambient temperature in hyperscale server rooms to increase efficiency and reduce the risk of overheating. CRAC units utilize refrigerant chemicals while CRAH units contain chilled water. Even in stable environments, this infrastructure can collapse, causing large spills.
Controlled Environments
Factors like temperature, humidity, CO2, and airflow can impact server performance. To ensure maximum efficiency, prevent outages, and avoid repair and replacement costs, hyperscale data centers need to maintain carefully controlled environments and receive immediate alerts to hazardous changes.
Leak Detection
Leaks are one of the leading causes of downtime in data centers. They often start small and escalate quickly to affect multiple areas and can result in facility-wide outages. Cooling systems are responsible for a majority of these leaks, and the high-density AI/ML clusters and GPUs in hyperscalers have especially demanding cooling needs.
Many hyperscalers have turned to in-server liquid cooling methods like direct-to-chip cooling, which relies on an elaborate system of tubes to deliver coolant fluid to servers. The complexity of direct-to-chip liquid cooling means there is a greater potential for leaks, and without an advanced detection system, leaks can be extremely difficult to locate and contain.
Parameter’s leak detection technology has evolved with the changing hyperscale landscape to encompass the needs of in-server liquid cooling. Powered by RLE Leak Detection, our proprietary server-level direct-to-chip liquid cooling leak detection system is already being implemented in critical data center operations. This technology can pinpoint the precise location of a leak as soon as it occurs, ensuring technicians have enough time to contain it before it spreads. Parameter solutions can also be used to detect water and conductive leaks caused by CRAC/CRAH infrastructure.
Environmental Monitoring
In the high-density world of hyperscale data centers, maintaining an optimal environment and keeping temperatures low is pivotal for preventing outages and overheating.
When atmospheric conditions fall outside of expected parameters, hyperscalers can face far-reaching consequences such as efficiency losses, equipment damage, extended periods of downtime, and even serious biohazard risks necessitating personnel evacuation. Because atmospheric changes can happen suddenly and can’t be detected with the naked eye, hyperscalers need effective monitoring systems to ensure uptime SLA compliance.
Parameter empowers hyperscale data centers with visibility into critical aspects of environmental infrastructure. RLE Environmental Monitoring technology tracks multiple conditions, including temperature, humidity, airflow, CO2, and more. When a failure is detected, the system immediately sends out an alert indicating the nature and location of the issue.
For hyperscalers balancing multiple priorities, Parameter’s revolutionary WiNG wireless line brings convenience to the forefront. The equipment is as easy to install as placing a magnet on a doorframe, and automates data collection as soon as setup is complete.
Battery Monitoring
For hyperscalers, protecting UPS assets is a necessary part of meeting uptime SLA commitments. Battery-powered UPS systems allow data centers to maintain uptime during grid outages, but they are also prone to failures. It only takes a single failed cell to compromise an entire string, and with so many loaded cells across multiple sites, the rate of failures is multiplied by the size of the operation.
Hyperscale data centers can ensure UPS integrity by implementing a comprehensive battery monitoring system that is semi-permanently connected to provide live insights into fleet health. Through Cellwatch, Parameter technology measures multiple conditions, alerts to failures, and delivers live data on critical units.
Parameter’s scalable Cellwatch Central platform is specifically designed for hyperscale applications. With a flexible and adjustable interface, it automates all stages of the battery monitoring process and can accommodate the intensive demands of large hyperscale UPS systems.
Product Lines
Ideal for high-density hyperscale applications, Parameter’s RLE Leak Detection product line prioritizes accuracy and responsiveness to accommodate the risks of in-server liquid cooling. To ensure full coverage, Parameter offers leak detection for conductive fluid and chemical leaks.
Liquid Sensing Cables
Parameter’s conductive fluid sensing cables can be run around the perimeter or a room or in a serpentine pattern under the floor for comprehensive coverage of server rooms with unmated accuracy and reliability. For refrigerant leaks, Parameter’s chemical sensing cables offer an especially fast reaction time and are especially durable and resistant to chemical corrosion.
Distance-Read Controllers
Parameter’s solution for hyperscale liquid cooling applications is the LD1500 distance-read controller, which monitors up to 1500 feet of fluid sensing cables. When mounted to each server rack and connected to flow valves, LD1500 controllers prevent server damage by immediately halting coolant flow, notifying personnel, and identifying the location of the incident as soon as a leak is detected. For large fleets of up to 128 LD1500 controllers, the feature-rich LD5200 controller can aggregate alerts and updates into a unified interface.
Spot Detectors & Zone Controllers
For more basic applications and readily visible areas such as HVAC drip pans, floor drains, and condensation outlets, Parameter’s simple and cost-effective spot detectors and zone controllers notify personnel as soon as a leak is detected in a specific area.
With built-in security and remote monitoring capabilities, Parameter’s RLE Environmental Monitoring platform is designed to help hyperscalers sustain uptime and prevent server failures by tracking critical environmental factors. Wired, wireless, and Wi-Fi equipment options ensure streamlined automation and integration across multiple sites.
WiNG Monitoring
Parameter’s hyper-convenient and reliable WiNG wireless platform features sensors for dozens of environmental conditions and automatically relays continuous data to the WiNG Manger, allowing hyperscalers to meet efficiency goals, support continuous uptime, and optimize operations. There are no recurring fees, and once you’ve purchased the equipment, it’s yours for life.
Wi-Fi Sensors
Parameter’s Wi-Fi sensors enable hyperscalers to leverage existing Wi-Fi networks to consistently and accurately monitor conditions such as temperature, humidity, CO2, and leaks. Our Wi-Fi sensors are compatible with WiNG-MGR or BMS systems.
Wired Facility Monitoring & Sensors
For operations that require the utmost security, Parameter simplifies closed-circuit monitoring through the wired Facility Monitoring System (FMS), which monitors sensitive facilities, temperature, and equipment status. To expand FMS functionality, Parameter offers wired sensors for temperature, humidity, motion, door status, smoke, power failures, and power currents.
Cellwatch Battery Monitoring is Parameter’s tailored solution for almost all UPS applications in hyperscale data centers. Cellwatch Battery Monitoring utilizes advanced measurement techniques to continuously monitor entire systems and provide extremely accurate readings throughout charge, discharge, and float periods. The platform integrates seamlessly with native systems to automate data collection and manage fleet health.
Real-Time Fleet-Wide Visibility
Cellwatch Battery Monitoring is semi-permanently connected to provide 24/7 automated monitoring of entire UPS fleets. It alerts to microfailures as soon as they occur, allowing personnel to safeguard UPS integrity and maintain uptime by replacing faulty cells before they need to be deployed.
Plug-and-Play Installation
To enhance automation and integration across hyperscale sites, Cellwatch Battery Monitoring is designed for easy, fast, and safe plug-and-play installation with identical modular components for all configurations. Cellwatch also facilitates streamlined Modbus SNMP integration.
Cellwatch Central
In addition to the core Cellwatch product, Parameter also offers Cellwatch Central, a specially tailored platform for multisite monitoring. By consolidating data collection into a unified system, Cellwatch Central enables colocation to analyze battery performance across over 50,000 jars and streamline management of battery types, installation data, and maintenance events via a unified system.
Discover Advanced Data Center Monitoring Solutions From Parameter.
Parameter monitoring systems are specifically designed with scalability, automation, and integration in mind to support the unique requirements of hyperscale data centers.
Contact Parameter to discover flexible monitoring solutions for hyperscalers that enable incident prevention, maximize uptime, and deliver peace of mind.
For custom capabilities and manufacturing, reach out to our OEM team directly to discuss how we can augment your application.