Data centers have evolved significantly in the last few years, becoming the central hub for almost all technologies. Whether we are talking about content delivery, business support, or data distribution, the data center is crucial for today’s business to succeed.
We can achieve an effective data center operation through a balanced investment in the facility and the housed equipment.
DATA CENTER DESIGN
As a result of IT equipment’s complexity within a data center, the availability, reliability, and maintenance of its parts demand more attention.
Good data center planning should provide a resilient environment. It is essential to evaluate the data center’s power and cooling requirements before building or upgrading. The plan should include the best ways to provide backup power if faced with an outage and how long it can sustain uptime during the blackout.
Servers, networking equipment, and appliances in a data center generate immense heat while working. A proper data center cooling design will consider the equipment’s location to prevent hotspots.
Data centers are one of the world’s most energy-consuming facilities because they need to provide access to information 24/7. As a result, the data center design should consider ways to optimize space and keep equipment within specific temperature and humidity ranges.
An accurate overview of the equipment installed in the data center allows us to plan the electrical work. Applications don’t react well when the power is interrupted. To prevent power outages, we need an uninterruptible power supply (UPS). When the power goes out, the UPS will switch over the current load to a set of internal or external batteries.
According to Hexa Research, the U.S. economy loses up to $550 billion each year due to power outages and other disruptions. Thus, thorough planning is needed to ensure the proper UPS is purchased, installed, and correctly managed.
With electrical power comes the inevitable dispersion of heat; the more power consumed, the more heat is expelled. The data center cooling design can address this heat concern by installing in-row cooling, overhead cooling, raised floor with underfloor cooling, and wall-mounted cooling.
A proper data center floor design will allow for easy growth. Once we finish evaluating power and cooling, we’ll start installing racks or cabinets.
The main considerations for a data center include:
- Core components – including computing units, data storage devices, and other hardware such as switches and routers
- Cabling System – cables connecting all the machines on-premise in a data center, including power, networking, devices, and other resources
- Uninterruptible Power Sources (UPS) – battery banks, generators with fuel storage, and redundant power sources
- Environmental control – computer room air conditioners (CRAC); heating, ventilation, air conditioning (HVAC) systems; and exhaust systems
- Fire Protection – smoke detectors, fire sprinkler systems, and passive systems like resistant barriers
- Physical security systems – biometrics and other surveillance systems
Data centers play a significant role in storing and securing a business’s critical data in the current IT environment. As data centers have become vital for enterprise IT operations, it is equally important to keep them running efficiently.
Here are the top five things we can do to sustain our data center’s reliability and efficiency.
STOP THE DAMAGE BEFORE IT BEGINS
When it comes to data centers, core components must be kept in good working order. The “fix it when it breaks” strategy is not a good approach for assets such as storage systems, servers, backup generators, CRAC units, or fire suppression systems. Datacenter organizations should not endorse a reactive mentality for systems support systems such as HVAC, UPS, and generators.
SCHEDULE REGULAR MAINTENANCE AND INSPECTIONS
Preventive maintenance on UPS and batteries diminishes the risk of failure during power outages. Emerson Network Power has published a study showing that the Mean Time Between Failures (MTBF) for UPS units that received two preventive maintenance service events in a specific year is 23 times higher than that of a UPS machine with no PM service events during that year. Preventative maintenance can reduce the chance of failure for every other critical system. Also, it will reduce the amount of energy consumed and will extend the equipment lifetime.
MAINTAIN DETAILED RECORDS
Record-keeping is essential in any data center because we need ample information during troubleshooting, downtime events, and other emergencies.
- Checklists – offer consistency and completeness in carrying out tasks. Checklists help us ensure no steps, procedures, or processes are missed out.
- Reports – should include regular maintenance dates, unexpected failures, and any reported issues. They are also essential for auditing requirements.
FOLLOW REGULAR MAINTENANCE ROUTINES</h3
Do not overlook regular maintenance. Using predictive technology in concert with periodic preventive maintenance reduces the number of reactive interventions and subsequent downtime.
Here are some things to check during each maintenance routine:
- Equipment – for performance, optimization, dust, heat, and so on
- Cables – to prevent any trip hazards and static build-up
- Floors/walls/ceilings – for potential leaks, cracks, air cooling loss
- Cooling systems and accessories – to ensure air conditioning units are clean; inspect blanking panels and cooling trays
- Power-related systems – to verify all power-related systems, including central, backups, distribution, etc
- Fire and other hazards – to check all sprinklers and other systems, confirming we can efficiently use them in case of an emergency
PERFORM PREVENTATIVE MAINTENANCE
Unplanned maintenance usually costs 3–9 times more than planned maintenance, mainly due to overtime-labor costs, collateral damage, emergency parts, and service call-outs. A Ponemon study revealed that a complete unplanned data center outage averages about 2.5 hours, costing almost $700k per incident.
Environment factors such as moisture and humidity could corrode different components leading to power failure. Data centers can ensure they are running efficiently by performing preventive maintenance. Servicing the needed parts and upgrading them can increase life and the data center infrastructure’s overall efficiency.
Tracking and measuring are key pillars of a successful business. So measuring the performance of preventive maintenance is essential. Maintenance tracking software simplifies the process of dealing with maintenance issues. We can use Maintenance Management Systems to help ensure maintenance is scheduled and performed routinely. We can also archive the maintenance history, track effectiveness, and build a digital database of the entire inventory to match against OEM or third-party maintenance (TPM) contract data to find and optimize unnecessary costs. Did you know that TPM can lower ongoing maintenance expenses even after the OEM warranty ends?