Four ways to maintain high availability of infrastructure

With planning and creativity, you can significantly simplify the task, without spending too much

Ensure the safety and availability of infrastructure would be a much simpler task if all companies have budget left to technology: it would be enough to invest much money in the most advanced systems redundancy, completely comprehensive.

Unfortunately, most companies can not afford it. But some small steps can significantly simplify the recovery process in the event of various types of problems. Some tips help in this task.

1 -Keep replacement items for each equipment
An ideal infrastructure is standardized and maintains servers and switches of the same brand, completely compatible with each other. Based on this principle, keep a replacement item for each machine can save a lot of headaches. A server, for example, can be replaced in a matter of minutes with a simple transport pucks. For routers and switches, worth the same principle: have configured equipment for quick replacements without impact on the activities of the organization.

Obviously, you can not keep spare some very expensive equipment such as core switches, but with a more focused maintenance on key equipment the rest is easy to cover with little budget.

2 – Use wiki for infrastructure inventory
What is the serial number of that switch to the remote office? What operating system version was running on that server before the power supply burned? It may be easier to organize this information in a wiki, installable on a virtual machine. With information available quickly, the recovery, which could take hours, it can take less than 30 minutes.

Start this project can be challenging. But once implemented, maintaining becomes easier. Remember this the next time you spend a lot of time by not obtain the serial number of a component and have trouble because of that.

3 – Keep backup links where and when possible
Where possible, the organization must maintain multiple paths for each data center and remote office. This was very expensive, but today it is possible to maintain alternative approaches for less than 200 dollars a month, using DSL solutions. Of course that will not be as fast as traditional connections, but much help in emergency situations. The key is also creativity.

4 – Have an emergency shelter for critical virtual machines
This tip applies to virtualized infrastructures. Let’s say the organization has a dozen servers running hundreds of virtual machines. If something goes wrong with the production system, the team can get away with when transferring virtual machines problematic server, keeping the line of critical applications in order. In this case, it is not necessary to duplicate the entire infrastructure, since it is possible to invest in only one server with four CPUs with sufficient RAM to ensure production loads for a period of time.

It is good to note that the size of the server must match the size of the total infrastructure of the company. Looking from this angle, it is not expensive to keep a machine capable of hosting virtual machines that run out eventually host.

Full article: http://cio.com.br/

Comments are closed.