Environmental Monitoring & Control
July 30, 2020
Data center and server room monitoring and control is essential in powering our modern world. It is the life support upon which our digital lives depend. However, when a server room isn’t properly managed it can fail, and the consequences can be extreme. It’s essential to carefully monitor the server room environment with dependable equipment you can trust.
Here are 7 important environmental monitoring and control functions you’ll need to better manage a server room:
Perhaps the most important task is to monitor and control the room temperature. There are two big reasons why:
Heat = Failure
Overheating is a sure way to take down servers and could result in the loss of huge volumes of data. Servers are in danger of overheating when room temperatures rise above 82° F. Cooler is better, within reason, but striking the right balance is key.
Server rooms use a lot of energy and keeping them cool is a primary contributor, adding substantial operational cost. While a server room would love a chilly 50° F, the average temperature falls somewhere between 73° and 75° F. Google famously runs their data centers at 80° F to conserve energy.
A temperature sensor and control system can be used in a variety of ways to enhance both server room performance and energy conservation. The obvious way would be to signal the primary HVAC system to turn on and off. However, there are important auxiliary functions that benefit many server rooms, especially as needs fluctuate.
Auxiliary fans, portable air conditioners, and additional vents can all be added to improve cooling when the primary HVAC system is not enough. This can happen as more servers are added, when ad hoc server closets are built, or as server rooms are expanded.
Both primary and auxiliary cooling measures need a control system to automate their operation. It all starts with temperature monitoring and a controller with the right I/O and built-in logic to intelligently start and stop each function.
Another critical parameter is humidity. While often overlooked in small installations, like a server closet, it is always important to monitor this essential environmental factor.
There are two risks from humidity:
- Low Humidity – Results in the buildup of electrostatic energy and could result in a static discharge which damages hardware and causes data loss.
- High Humidity – Leads to condensation and corrosion of electrical components.
Both high and low humidity conditions need to be avoided to ensure the longevity of the equipment and ultimately, the integrity of the servers’ data.
Optimal humidity in your server room lies between 40% and 60%. If your humidity level is falling outside of these parameters, your monitoring and control system needs to automate the right process and alert you to the issue. Sometimes your HVAC system will fail to regulate humidity. This indicates the need for service, or for the addition of a humidifier or dehumidifier, depending on your situation. For this reason, having both automated control and status alerts from your environmental control system is important.
ControlByWeb probe which monitors both temperature and humidity in one probe.
3. Leak Detection
The modern data center increasingly relies on liquid cooled server racks to help regulate temperature. These units can drastically reduce cooling costs and for this reason are becoming more common in large and small server rooms. While these systems are effective solutions to an ever-present problem, they also introduce a greater risk of leaks.
Additionally, leaks in HVAC systems are common and can enter the server room environment through walls or from above with suspended AC units. Nearby plumbing can also leak, adding a layer of risk that should be mitigated.
Leak detection in your server room environment is a must. A leak monitor should alert the right personnel immediately upon detection to prevent a potentially major disaster.
4. Remote Reboot
Rebooting your server or other network devices is necessary when it becomes unresponsive, or for routine maintenance.
Performing a remote reboot on your server or other network device can be done in two ways:
- A soft reboot can be done virtually using command lines which relies on the software to execute the process. This is often sufficient for software updates and maintenance.
- A hard reboot is done by power cycling the server. This is required when a server becomes completely unresponsive.
Normally hard rebooting requires a technician to be on-site. However, with the right equipment you can perform hard reboots remotely.
A good example of equipment that lets you remotely perform a hard reboot is the WebSwitch. It allows both manual and automatic rebooting through a simple web interface which you can access in a standard web browser.
This is also done for IP security cameras which, just like servers, can become unresponsive. An automatic, hard reboot brings them back to life.
5. Door Status
Only those who are authorized and know what they’re doing should enter server room areas. Additionally, too much traffic can result in an excess of dust and particulates. An important step in controlling access is monitoring traffic with a door status sensor.
Did you know that most of the dust introduced into a server room is from human traffic? According to Dataspan, 80% of the dust and dirt entering your server room is tracked in on the soles of shoes. They claim that if not regularly removed, this dust can raise equipment temperature by as much as 30° F.
Add door monitoring and logging to help you protect your equipment, reduce cleaning, and avoid unnecessary maintenance.
6. Generator Fuel Tank Levels
Data centers rely on an ample supply of fuel for their backup generators. This fuel inventory can be monitored separately, or it can be integrated with your server room control system. The right liquid level monitoring technology, paired with real-time remote telemetry, is essential for smooth operation.
Additionally, complete generator monitoring can be included depending on the scope and goals of your monitoring and control system. Generators can be remotely started and stopped, power status and battery voltage can be monitored, and even the engine block can be monitored to ensure good working order.
You can learn more about generator monitoring and control here
7. Demand-Side Power Management
Your control system can help you make important demand-side optimizations. When a data center has a change in available power, such as when it switches to backup generators, it may be beneficial to power down unnecessary processes in favor of critical equipment. A good control system can monitor the generators’ operational status and begin demand-side power management.
For example, equipment can be broken into zones so the control system can selectively shut down each zone based on a user-defined priority. This provides more power for the servers themselves and their cooling systems.
Demand-side power management is beneficial as it reserves a limited supply for critical tasks, improves energy efficiency when it matters most, and it reduces heat production as a whole, easing the burden of temperature control.
Server Room Alarms
Alarms and notifications are important to notify staff of actions they need to take to prevent an emergency. Remote alarms are especially helpful as data centers and server rooms are increasingly managed remotely. Utilizing these alarms for your data center takes the guesswork out of managing server rooms.
Trigger Local Alarms
Local alarms of all types, from audible sounds to flashing lights, can alert on-site staff of leaks, doors left open, rising temperatures and more. These alarms don’t rely on 3rd party network connections to work and are great for providing redundancies for texts and email alerts.
Send Email & Text Message Alerts
Perhaps the best way to be notified of a noteworthy condition in your server room or data center is through sending email and text message alerts. These alerts can reach your personal cellular device, keeping you aware of important changes at all times, no matter where you are. Email and text alerts are extremely valuable and have prevented many disasters in all kinds of industries throughout the world.
Balancing Cost & Benefits
How you implement server room monitoring and control in your data center or office ultimately depends on your specific situation. These 7 key parameters are critical pieces which should be considered thoroughly.
The good news is you can do all this and more with a cost effective, industrial solution using web-enabled Ethernet I/O devices from ControlByWeb. In fact, Comparitech recently rated us in the top 10 environmental monitoring solutions for server rooms.
Our I/O devices excel in server room monitoring and control, including basic and advanced solutions. Unlike some off-the-shelf environmental monitoring packages, our industrial-grade devices work with most monitoring and control equipment and communicate using a variety of common protocols like SNMP, XML, and Modbus TCP/IP. This allows you to create a custom solution to meet your exact needs.
Importantly, our devices do not require a cloud or any subscriptions, so you can enjoy full functionality without additional fees. You can access your control system either on your local IP network or via the Internet.
Check out our case study on server room monitoring to see how we can be a part of your solution.
Credit to Control By Web.
Water and Wastewater
Many companies in the water and waste-water sector use Variable Frequency Drives (VFDs) to efficiently control the electrical power needed to run motors, pumps, and other mechanical systems. While the VFDs provide greater control over the output, they also have a tendency to introduce harmful harmonic distortion into your electrical system. When this occurs, you can experience overheating of components, equipment malfunction, nuisance tripping of circuit breakers, and inaccurate measurements on sensors.
In the water industry, it is quite common to see the use of multi-pulse drives to reduce harmonic distortion. However, there are better options. MTE has worked with a number of companies in the water industry to control harmonics using a six-pulse drive with the Matrix® AP passive harmonic filter. In addition to helping them meet IEEE-519, standards, the solution is smaller, more efficient and more cost effective over other low harmonic drive solutions.
The 6-Pulse drive with Matrix AP harmonic filter outperformed the 18-pulse drive in the following important areas*:
Power loss: 40% – 73% depending on the size of the VFD and the component manufacturer.
Overall efficiency: 0.5% more efficient (99.0% vs. 98.5%).
Harmonic performance under balanced line conditions: 1-2% better THID performance for loads 25-75% and equal performance for loads greater than 75%.
Harmonic performance under line imbalance conditions: Significantly better performance. For example, under 3% line imbalance, performance was 32.5% better (17.5% vs 50% THID) at 25% load and 13% better (12% vs 25% THID) at 50% load.
Power Factor: Better to equal performance for loads 50-100%.
*These findings are based on the performance of a 100 HP 18-pulse drive with the performance of a standard 100 HP 6-pulse drive equipped with a 128 amp Matrix AP harmonic filter.
Credits to MTE