Creating an IT Infrastructure for Business Continuity
Business continuity is an issue that no organization can afford to ignore. In fact, according to The Definitive Handbook for Business Management, between 60 and 90 percent of companies without a proactive disaster plan find themselves out of business within 24 months of experiencing a major disaster.
Increasingly, IT power and cooling are becoming more important as factors in ensuring business continuity. These 10 steps provide a good start for ensuring the integrity and availability of your IT systems.
1. Assess your situation.
Review existing power and cooling systems to identify threats and vulnerabilities to business continuity.
2. Ensure the physical security of your equipment
While large data centers often have strict access policies and procedures, smaller locations or more remote locations may not. It’s important to use racks that come with key or card swipe locks and contact closures that protect against unauthorized access. These locks and closures can be connected to your network so you can easily provide authorizations and monitor access. Within the rack, smart PDUs enable control of individual receptacles. This prevents unauthorized equipment additions that can overload circuits and create a power outage.
3. Keep your cool
High heat can reduce the performance of equipment. IT equipment often requires 24×7 dedicated cooling, precise temperature, humidity and air filtration control and more efficient cooling provided only by precision cooling. Typically, racks with 1kW to 3kW need dedicated cooling, either through single cabinets with integrated cooling or through room level cooling.
4. Eliminate hot spots
Many data center now have high density servers. These can create hot spots, causing equipment degradation and under utilization of rack space. At 5kW and above, high-density cooling often is required. You can resolve hot spots in single racks with cabinets featuring integrated high density cooling or in multiple racks by using high density supplemental cooling in areas already served by room level cooling.
5. Ensure power quality
The most commonly used UPSs are line-interactive and do not condition certain power problems such as frequency variations and distortions until they go to battery. They may pass utility power irregularities to the protected equipment, resulting in the power being dropped. You can mitigate this risk by using online UPSs, which fully condition utility power before passing it to the protected equipment.
6. Get the UPS capacity you need
Insufficient runtime is the second most common cause of UPS failure. Ensure that UPS sizing and backup are adequate for your current environment and for future growth. Calculate UPS size based on the full load of protected equipment – not on “nominal loads” which are estimates of average loads and could result in undersizing UPS capacity.
7. Increase UPS reliability
Commonly used line-interactive UPSs will drop power if a component fails. Online UPSs are twice as reliable, as measured in MTBF. That’s because they has an internal bypass that allows power to continue to the protected equipment in the event of a UPS component failure. If total room power load exceeds 15kW, it may be time to replace multiple rack UPS systems with a room level system, which reduces the potential points of failure created by the multiple UPS systems.
8. Add redundancy
Dual corded network equipment is designed for redundancy – two PDUs, two UPSs, two power circuits – to protect availability in case a single component in the power chain fails. Redundancy down to the dual corded load, not just the UPS, is required to maintain highest levels of availability for critical loads.
9. Ensure visibility and proactive monitoring
The power and cooling equipment your IT systems depend on can be configured with Webcards that enable SNMP monitoring of the IT infrastructure over the existing network. If monitoring of critical systems is too time consuming, consider outsourcing remote monitoring. Environmental conditions – temperature, humidity and water leakage – also must be monitored.
10. Have a Strategy for Service
Extending the useful service life of the power and cooling equipment through proper maintenance, predictive monitoring, and keeping the equipment up to date increases equipment lifespan and maximizes performance. Be sure to use factory-certified service technicians who can work with your local solutions provider to provide rapid response and continuous maintenance.
eMazzanti Technologies can help you assess your IT spaces and determine appropriate power and cooling solutions for business continuity. See www.emazzanti.net for details.