Colombia. The mass migration of critical applications from traditional data to the cloud grabs the attention of analysts, industry observers, and data center stakeholders. However, as this migration transforms the data center industry, a smaller and less noticeable revolution has been taking place around non-cloud applications that have been left behind. These "Edge" or edge applications have remained on site and, due to the nature of the cloud, the criticality of these has increased significantly.
Schneider Electric, a leader in digital transformation of energy management and automation, explains factors to consider when implementing these solutions. The centralized cloud was conceived for applications where time was not absolutely crucial, but critical applications shifted to the cloud and it became apparent that latency, bandwidth limitations, security, and other regulatory requirements were putting limits on what could be placed in it.
Therefore, it was considered -case by case- that certain existing applications (for example, the processing of factory floors), and in fact some new emerging ones (for cars, smart traffic lights and others associated with the Internet of Things that require high bandwidth), remained in the "Edge".
Considering the nature of these rapid changes, it is recurrent that some data center planners misinterpret the cloud trend and equip on-site data center capacity with lower criticality. Due to the need to achieve a higher level of control, adherence to regulatory requirements, decrease latency and improve connectivity, these new Edge data centers need to be designed with criticality and high memory availability.
The problem is that many small-site data centers are not designed to take on their new role as solutions for critical data. Most are organized as one or two servers housed within a wired closet. As such, these sites, the way they are currently configured, are prone to unscheduled system downtime or downtime and physical security risks – therefore – require some rethinking.
On the other hand we find the redundancy of systems, which is also a conflict with most applications that are hosted in the cloud, because when the access point is down employees lose productivity. Edge technology, by staying up and running during these downtime scenarios, helps bolster business continuity. Schneider Electric recommends the following practices to improve critical application availability:
Enhanced security: When entering some of these server rooms and cabinets, you typically see unsecured doors and open racks (no doors). In these cases, the equipment must be moved to a closed room and at best implement biometric access control. For harsh environments, equipment should be secured in an enclosure that protects against dust, water, moisture and vandalism with video surveillance, adding 24 x 7 environmental monitoring.
Dedicated cooling: Small rooms and traditional closets often rely on the most comfortable cooling system the building has. This may no longer be enough to keep systems running. Reevaluate the system and humidification to see if it's appropriate, as a passive airflow, another active airflow, or a dedicated cooling approach is now needed.
DCIM Management: These rooms are often left alone without dedicated staff or software to manage assets and ensure downtime is avoided. Take inventory of existing management methods and systems. Consolidate into a centralized monitoring platform of all assets in these remote sites, with that you can maintain control when human resources are limited.
Rack management: Managing cables within racks in these remote locations is often an afterthought, causing cable clutter, obstructions to airflow within this system, and increased human error during additions/movements/changes. Modern racks, equipped with easy cable management options, can reduce the risks of unanticipated downtime.
Redundancy: Power systems (such as UPS) are typically the No. 1 solutions in traditional environments for keeping systems running when maintenance is performed. Consider redundant power paths for simultaneous maintainability at critical sites; make sure that the circuits are in the emergency generator, analyze whether the possibility of adding a second network provider for critical sites is necessary; organize network cables with management devices (channels, routing systems, and moorings), with color-coded labels and network lines to prevent human error.
To conclude, it is necessary to have a systematic approach to evaluate small remote data centers and thus ensure the highest return on investments in the Edge.
Leave your comment