The tip from last week covered the syntax for co-processes, and also used a short script to demonstrate how to start, write to, and read from a co-process.  If you missed it, you can view it here.

The objective of this week's tip is to get you thinking about what processing benefits can be gained from using co-processes.  Consider the following scenario...

You have a single program on a "master" UNIX or Linux server that needs to run once for every UNIX and Linux server in your environment.  When this program runs it collects data unique to each host that will be stored on the "master" server for additional processing.  This processing will be performed as soon as the data collection has completed for all servers.

If there was only a handful of systems in your environment you could just run a for loop within a shell script that runs the collection program against each host in a serial fashion.  If you had hundreds or thousands of servers in your environment, you would quickly discover that this would not be an acceptable solution if you had time constraints for the processing.

Think about how co-processes could be used to provide a more desirable option, and next week's tip will contain a potential solution.