Page tree
Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 4 Next »

What is the performance overhead of MoSKito

Well, the answer to this question is rather unsatisfactory: "Depends!". 

It depends on multiple values, first of all integration type and number of monitoring points. To be clear: each monitoring point costs time. How much time depends on integration type. For example Proxy integration is slower than AOP compile time weaving. However, in almost all cases the costs are in lower microseconds area, how low depends on your hardware. 

For most web applications the performance drawback of MoSKito will be not measurable. It may consume a bit more CPU but will be most surely unnoticed by the end user. 

In other words, you can use it without thinking about performance impacts, until you really see some. 

 

What is the memory footprint

MoSKito Essential runs within your app's heap space and therefore consumes some of it. Usually the amount of memory used by MoSKito is significantly lower than the amount of memory used by the monitored code. So in case you have millions of monitored classes and your heap size is above 10 Gb, MoSKito will consumer some few hundred MBs, which is about 1-3 percent of your heap. If you have a small application, with 128 MB heap MoSKito will usually consume 1-3 MB. You can calculate your exact memory footprint by creating a histogram. 

 

 

  • No labels