There are plenty of misconceptions and myths surrounding the data center infrastructure management (DCIM) market today.
We’d like to “slay” a few that have been rearing their ugly heads lately.
- Real time monitoring is all you need. While real-time monitoring is an important part of your data center’s operations, because it shows you what is going on “right now” and can help head off disasters before they happen. It’s also essential for insight into short and long-term trends. With trend analysis, patterns will likely become obvious, making long-term capacity planning and operational cost savings simpler. In addition, long-term information also shows the potential fragility of systems that may lack redundancy, and can help manage assets and the processes surrounding them better.
- Using a colocation facility means that you no longer need to manage resources. As the colocation facility keeps the lights and fans on, you’re still responsible for your assets and coordinating people and processes, ensuring that staff have access to specific assets, and that the colo meets or exceeds your service level agreement (SLA).
- The public cloud is less expensive. This cloud cost statement is not always true. Sometimes the public cloud is simply more expensive, and there can be plenty of hidden costs. For example, the cost to convert, implement and integrate applications can be significant. Some applications can be very costly to run in the public cloud if they are database intensive or have other unique requirements. Plus the extra costs for premium services (data transfer, etc.) add up quickly.
- All resources will be run in the public cloud. Actually, the hybrid model is becoming more and more popular. There are several reasons for this:• Some legacy applications are difficult to maintain in the public cloud.
• Compliance sometimes prevents public cloud use in certain industries or regions.
• Performance can be an issue for some types of applications.
• Some organizations don’t want to become completely dependent on an outside supplier.
- Having a CMDB means you have control of your assets. CMDBs don’t manage workflow or processes in either an enterprise or a colocation facility. They also quickly become out of sync with what’s really going on in the data center, and affect everyone in the organization who depends on accurate information.
- If you’ve planned for outages, you’re protected. Planning for outages is responsible and important. But failure simulation, to see what critical systems would potentially be affected if an outage occurred, should also be performed. Running planned outages is ultimately the only way to see what would really happen in the event of a failure.
- Virtualization means you don’t have to worry about hardware any more. No doubt that virtualization of your hardware improves operational performance. But it doesn’t change the operational efficiencies at the physical infrastructure layer. Asset, capacity planning and power management are still required. In fact, increased density at the VM layer causes increased resource utilization and risk at the physical level, and requires closer management.
Don’t buy into myths surrounding DCIM. As the world of hybrid infrastructure grows, organizations that automate and change their management approach to keep up with architecture and maintain control will be most successful.