Cloud computing Management three great tools: Nagios, Ganglia and Splunk

Comprehensive utilization of Nagios, ganglia and splunk cloud computing platform monitoring system, with error alarm, performance tuning, problem tracking and automatic generation of operational dimension report function. With this system, you can easily manage the Hadoop/hbase cloud computing platform.

Cloud computing has long been not a conceptual phase, with large companies buying a large number of machines to begin formal deployments and operations. And the performance of hundreds of powerful servers, for operational management has brought great challenges.

If there is no convenient monitoring and alarm platform, the administrator is like a nightmare, every day will be like firefighters, the rapid tapping of the keyboard, with the original UNIX command in many machines.

If there is no good log management platform, for developers troubleshooting is a tearful thing.

And if you're the head of the team, it's important to be concise and clear. Stakeholders are likely to ask the system SLA, machine utilization and many other problems, after all, the company invested a huge amount of money and manpower.

Friends, what do we do when we manage the cloud computing platform that our company has high hopes for when we face so many real challenges?

Overview

When we build the trend cloud computing platform, we encounter many problems and challenges. At the beginning of the construction, the first time there are so many powerful machines, we feel excited at the same time, there are some concerns. Everyone sat down to discuss the problem with a full whiteboard.

What if there is a problem, is there an early warning mechanism?

Is there a visual management interface?

Does the management platform need to be developed by itself? How difficult is development?

Are there open source management tools?

So many logs distributed on each machine, there is no more effective way to manage?

Can I generate a good report?

Machine downtime, can the administrator receive SMS notification?

How do I do performance tuning?

Can you give a basis for expansion and upgrade?

With these questions, we started our own cloud computing platform management and operation of the journey, all the way, the harvest is abundant. Now basically formed a set of cloud computing platform monitoring system as shown in Figure 1.

Figure 1 Cloud computing Platform Monitoring architecture

In this system, we use the Nagios, ganglia and splunk, build a cloud computing platform monitoring system, so that it has error alarm, performance tuning, problem tracking and automatic generation of operational dimension report function. With this system, we are finally able to easily manage the Hadoop/hbase cloud computing platform. The next step is to briefly describe their features and capabilities.

(Responsible editor: The good of the Legacy)