using Http Event Collector, splunk loggin driver consumes too much CPU: 94-100%
anyone knows what's the reason behind it and how to det it down?
Short answer - your CPU is at constant 100% because your system is overutilized.
Long answer - most probably you're sending more data than the host can handle. Check your pipelines, check your inputs and outputs. Check your memory usage. There are so many things that can be wrong here.
Your question is about as detailed as "my car won't go when I press the accelerator pedal" when we don't know if you're out of gas, the battery is dead or whether you simply forgot to put it in gear.