So I've been integrating Sigar and its turning out fairly well, and I have come to a simple question: How to best use the Sigar object? Let me explain:
For each system attribute being monitored there is a corresponding component that is creating a new Sigar() object. So I have one for CPU, one for System Memory, one for Process Memory, etc ...
I am wondering if its a good idea to use a new Sigar() for each of these components, or whether SigarProxyCache.newInstance(...) should be used, creating one Sigar() upfront and each component use the SigarProxy to access system information.
I suppose this is based on how expensive creating a new Sigar() is. What do you recommend?
Also, it seems that only way to modify the periodicity that system metrics are collected is to use SigarProxyCache.newInstance(...), in general, is this the recommended way to use the Sigar object?
Thanks
Dennis
For each system attribute being monitored there is a corresponding component that is creating a new Sigar() object. So I have one for CPU, one for System Memory, one for Process Memory, etc ...
I am wondering if its a good idea to use a new Sigar() for each of these components, or whether SigarProxyCache.newInstance(...) should be used, creating one Sigar() upfront and each component use the SigarProxy to access system information.
I suppose this is based on how expensive creating a new Sigar() is. What do you recommend?
Also, it seems that only way to modify the periodicity that system metrics are collected is to use SigarProxyCache.newInstance(...), in general, is this the recommended way to use the Sigar object?
Thanks
Dennis