SIEMENS Data Centers Reference Architecture
INSTRUCTION MANUAL
A fault-tolerant DDC-based BMS architecture designed for data centers, providing redundancy and backup systems to ensure uninterrupted operations even in case of infrastructure failures.
The architecture includes components such as protection relays, switchgear, transformers, switchboards, controllers, meters, valves, sensors, and more.
DATA CENTERS
Reference
Architecture | Fault-tolerant DDC-based BMS architecture for certified Tier IV
siemens.com/datacenters
Introduction
A reference architecture is a standardized blueprint that outlines the components, communication protocols, and functional relationships between the different systems and devices in a data center. It provides a unified framework to ensure seamless integration of various components, including those from third-party vendors. Furthermore, a reference architecture facilitates scalability and adaptability as a data center’s requirements evolve.
This flyer outlines one possible design, implementation, and operation of a data center’s monitoring, control, and management system – commonly referred to as the Building Management System (BMS) and Electrical Power Monitoring System (EPMS). By adopting this solution guideline, you will be in an even better position to enhance cost efficiency, reliability, and scalability
Reference Architecture Concept
The role of the BMS in redundancy
Downtime is the worst nightmare for data center operators. For a fault-tolerant data center, standards bodies like the Uptime Institute prescribe multiple backup systems and pathways to ensure continuous power and cooling – even if a part fails.
The Building Management System (BMS) must be able to align with this robust setup.
The core idea behind a fault-tolerant data center is to have backup systems in place. Simply put, double the needed capacity is provided, known as 2N. If part of the infrastructure fails, like cooling or power, the data center will still operate normally as the duplicate system takes over to keep everything running without interruption.
Take the example of an HVAC system in a data center. The Direct Digital Controller (DDC) for the HVAC typically consists of a central controller, remote sensors, actuators, and communication networks. A single power supply can be used to monitor these field devices and cooling units, because the overall system’s redundancy ensures continued operation in the event of a failure. The redundancy of the controllers is also not required due to the redundancy at the system level
In short, if a DDC or its controlled device(s) is not working properly or is faulty, the availability of the infrastructure remains in place and a maintenance task is triggered to fix the issue.
But what does this mean for the BMS? In such a scenario with system-level redundancy (2N) monitored by DDCs, two independent BMS applications have to be operated in parallel. On each server a BMS application oversees the physical infrastructure, maintaining optimal mechanical and power performances – as long as safety and security conditions for workers and visitors.
Any failure affecting one of the two BMS applications will not affect operations, due to the independency of the systems. In the case of the HVAC, the DDCs provide real-time data and control actions to both BMSs. If a BMS experiences a failure, the other takes over seamlessly and downtime is prevented.
Reference architecture schematic
Report from the field: BMS certification
To receive certification from the Uptime Institute, data center operations are subjected to a rigorous evaluation process.
Certification proceedings test a data center’s reliability, especially for Tier IV compliance, the Uptime Institutes most stringent rating. Among others, operators have to prove that their data centers can maintain essential power and cooling, even during maintenance. They have to show that the failure of a capacity system, component, or distribution element won’t impact computer equipment. And they are called on to demonstrate how their system can automatically handle failures.
The test list typically includes more than a hundred different scenarios. Here’s a sampling:
- If BMS-A malfunctions, BMS-B takes over monitoring and controlling critical IT equipment and cooling systems (and vice-versa).
- In the event of a BMS-A power panel failure, BMS-B reroutes power to critical IT and cooling systems using engine generators and DRUPS (Distributed Redundant Power Systems).
- Similarly, if the BMS-B power panel fails, BMS-A will utilize engine generators and DRUPS to maintain power for critical IT and cooling.
- A single BMS system failure (either A or B) won’t disrupt cooling operations within critical rooms.
- Facility staff can manually start, stop, and control the cooling plant even if the BMS is unavailable, ensuring uninterrupted cooling for critical systems.
Consistent with the robust redundancy strategy, two identical servers form the foundation of the BMS. This deployment ensures that any malfunction or outage affecting one server will not compromise data availability, alarms prioritization, and notification management. This dual-server setup allows us to keep data centers operating continuously and efficientl
Advantages of implementing compliant BMS architecture
The process requires careful planning, design, and execution to ensure adherence with the standards and fulfillment of all the test scenarios set by the certification authority, such as Uptime Institute. The benefits include:
- Enhanced Credibility: Certifications bolster the credibility of the data center operator, demonstrating to potential tenants that its infrastructure meets the highest industry standards of reliability and compliance.
This can significantly increase the operator’s attractiveness to demanding clients who prioritize data integrity and uptime. - Risk Mitigation: Certifications help data center operators mitigate risk by ensuring that their infrastructure meets industry best practices and is subject to regular audits. This proactive approach helps prevent equipment failures, downtime, and potential data loss, protecting the operator’s reputation and safeguarding their clients’ valuable information
- Process Optimization: The certification process itself can serve as a catalyst for process optimization within the data center. By aligning their operations with industry standards and undergoing thorough audits, data center operators identify areas for improvement and implement best practices, leading to enhanced efficiency and reduced operational costs.
Siemens offerings that support this reference architecture:
Desigo CC
Siemens’ integrated building management system for high performing buildings has the power to increase comfort, efficiency, resiliency, and safety; acting as the technology backbone to leverage the full potential of your smart building infrastructure.
Desigo PXC
Our Direct Digital Controller (DDC) provide optimal control for high-performing automation of all facility operations, based on on four key pillars: scalable, smart, simple, secured.
The cost-effective solution minimizes deployment time and operational overhead while maximizing reliability, performance, and long-term value across the data center lifecycle.
SENTRON Powermanager
This software lets you monitor energy consumption in buildings and plants, keep an eye on the condition of their low-voltage power distribution and on the power quality, and generate reports for operational power management according to ISO 50001
SCALANCE
These network components form the basis for communication networks in industrial automation. They meet all requirements for highly efficient industrial networks and bus systems, whether for switching, routing, security applications, remote access or industrial wireless LAN.
Published by
Siemens Switzerland Ltd
Smart Infrastructure
Global Headquarters
Theilerstrasse 1a
6300 Zug
Switzerland
Tel +41 58 724 24 24
For the U.S. published by
Siemens Industry Inc.
3617 Parkway Lane
Peactree Corners, GA 30092
United States
Article-No: SI-FlyerRefArchI-WWEN-2025-02
Status: 08/2025
Subject to changes and errors. The information given in this document only contains general descriptions and/or performance features which may not always specifically reflect those described, or which may undergo modification in the course of further development of the products. The requested performance features are binding only when they are expressly agreed upon in the concluded contract.
© 2025 by Siemens AG, Berlin and Munich
FAQ
Q: What is the purpose of a reference architecture in a data center?
A: A reference architecture provides a standardized blueprint for integrating components and systems in a data center to ensure scalability, adaptability, and seamless operation.
Q: How does the fault-tolerant DDC-based BMS architecture ensure uninterrupted operations?
A: The architecture includes redundant systems that can take over in case of failures, ensuring continuous operation without downtime.
Q: How important is certification for data center operations?
A: Certification from organizations like the Uptime Institute ensures that data centers meet stringent standards for reliability and performance, especially for critical operations like Tier IV compliance.
Documents / Resources
 |
SIEMENS Data Centers Reference Architecture [pdf] Instruction Manual Version 2.4.0, Data Centers Reference Architecture, Centers Reference Architecture, Reference Architecture |
