IT Services : Cloud Computing : Infrastructure – Johannesburg
Fire detection and suppression
Smoke/Fire detection and gas deployment
Fire suppression is a requirement for modern data centers. While there is a large upfront investment, these units last for some time. The costs are therefore amortized over that period. Gasses come in "green" editions which are optional. In either case inert cases prevent fires by removing one critical element in the requirements triangle of any fire: heat, combustible material and most importantly - oxygen. This happens without any risk to human life.
Power Mitigation and Planning
For stable 24/7 delivery
Modern data centers rely on stable power. In South Africa this is critical. Not only can we expect to see power outages for extended periods of time, but we need to ensure that all the core subsystems that rely on power continue to operate. The key to providing this is redundancy. Redundant supply and redundant delivery are needed. We have rolled out these systems in various sites and seen what can happen. When water floods a portion of a server room from a faulty air-conditioning unit. When a UPS fails or when a power cord explodes due to short circuits within the cable. We know how to build systems that work, and that continue to run 24x7.
So where are the low voltage cables going?
Overhead routing or raised flooring?
Cable reticulation can follow one of broadly two paths: Under floor or overhead. This is a critical choice for your data center. There are arguments to be made for both. This is the primary design choice since it has implications for all other elements of the data center. With taller cabinets and an increasing requirement to fit more technology into a smaller space, there is a strong trend towards overhead cabling. We can help you work out the subsequent choices for all other elements and cost benefits for each.
Cooling and environmental monitoring
Data center cooling differs from the demands of regular office cooling. Robustness is critical in these deployments. Redundancy is required in cooling systems as well as careful monitoring of systems. If one air conditioning fails, there is little point in that just being ignored until the second unit fails as well. Instead notifications need to be sent out to emergency contacts to ensure that systems are returned to service as soon as possible. We assist with choosing the correct cooling systems, working with contractors as well as actually installing and testing the monitoring units that detect failures and send alerts.
TechNiche has extensive experience in maintaining and deploying data centers. Having worked in and maintained these environments for over a decade, critical lessons have been learned. These translate into better solutions, better monitoring and of course, less failure.