As cloud deployments become bigger and more complex, monitoring becomes more important
Monitoring and management of cloud-based production environments is often an afterthought.
Why? Perhaps there is so much to think about — security, cost management, and governance —
that monitoring and management fall by the wayside. Or perhaps IT believes it simply can use
clouds’ native tools.
Monitoring and management are important to any data center, whether in the cloud or not.
These tools react to the real-time data gathered from system operations — storage, compute,
applications, databases, and so on — as well as respond to trends within the data.
[ The InfoWorld review: Azure Machine Learning is for pros only | Get a digest of the day’s
top tech stories in the InfoWorld Daily newsletter. ]
For example, say the performance of the database is falling behind the requirements of the
applications. Cloud monitoring and management tools would notify the cloud admins of this
situation, and either the admin or an automated process can take corrective action, such as
launching more machine instances to increase the database’s performance.
However, the most powerful benefit of cloud monitoring and management is the ability to
watch trends. This means gathering many data points over time and drawing conclusions as to
what they mean. For example, spotting accelerated use of storage services would suggest
there could be imminent performance and capacity problems. From there, proactive corrective
action could be taken, whether automated or manual.
Now that you’re sold on the benefits of cloud monitoring and management, how do you go about
choosing the right tools? There are three basic criteria:
1、Your monitoring and management tools should be cloud-agnostic. They need to span all he
clouds that you’re using, both private and public. If you have a tool that’s native to one
cloud, that’s no good.
2、Make sure your tool can gather system data over time, and then make sense of the data. In
other words, it should have data analytics capabilities.
3、Your tools should be able to take automated corrective action. This means you can
preprogram responses to common issues, so your clouds become self-healing.