All four cores at 100%? We don’t see that kind of spike when we open some other application right? Is it the experience of everyone? Or am I wrong in this?
I mean, maybe. Imagine averaging the usage over a single instruction cycle. Whatever ran that instruction is using 100% of the processor for that time period.
That first data point is simply invalid. Ignore it. Monitoring software usually report some kind of statistic (mean, median, min/max, etc) taken from measurements over a period of time instead of the instantaneous value when the report is updated. But they can’t do that for the first data point when the application is launched because there’s no time period over which to measure it.
This is the real answer. Other apps might be bottlenecked by IO so the CPU doesn’t work as hard. Get faster disks, the CPU will see more use. Since top is so small that it loads into memory almost instantly, and has no need for further IO, the CPU is free to spin all the way!
People get worked up when CPU usage is high, but unless there’s a resource leak somewhere, that just means the computer is working at full efficiency.
It’s like when the boss walks in so everybody becomes visibly busy. Never want to look like you could be the couple pennies shaved off the budget when management bonus calculation time comes around.
Absolutely. That’s why race to idle is a thing. As soon as the CPU is fed cycles it races to complete the task as quickly as possible. When the program first launches and starts monitoring the only thing it knows is that it was just doing something. It hasn’t had a change to not do something. It’s now completed its race and now it’s able to relax for a second and idle.
It’s going to be an artifact of the data collection method. The sample size is extremely small and taken during the period where top has never stopped running.
All four cores at 100%? We don’t see that kind of spike when we open some other application right? Is it the experience of everyone? Or am I wrong in this?
I mean, maybe. Imagine averaging the usage over a single instruction cycle. Whatever ran that instruction is using 100% of the processor for that time period.
That first data point is simply invalid. Ignore it. Monitoring software usually report some kind of statistic (mean, median, min/max, etc) taken from measurements over a period of time instead of the instantaneous value when the report is updated. But they can’t do that for the first data point when the application is launched because there’s no time period over which to measure it.
Lim t->0 x/t =infinity when x>0
Math - not even once.
I paid for all the cores, I’m gonna use all the cores!
Unless bottlenecked somewhere else, we would want to see 100 % with everything that takes more than some milliseconds.
This is the real answer. Other apps might be bottlenecked by IO so the CPU doesn’t work as hard. Get faster disks, the CPU will see more use. Since
topis so small that it loads into memory almost instantly, and has no need for further IO, the CPU is free to spin all the way!People get worked up when CPU usage is high, but unless there’s a resource leak somewhere, that just means the computer is working at full efficiency.
Top revs the engine as a test /s
It’s like when the boss walks in so everybody becomes visibly busy. Never want to look like you could be the couple pennies shaved off the budget when management bonus calculation time comes around.
Absolutely. That’s why race to idle is a thing. As soon as the CPU is fed cycles it races to complete the task as quickly as possible. When the program first launches and starts monitoring the only thing it knows is that it was just doing something. It hasn’t had a change to not do something. It’s now completed its race and now it’s able to relax for a second and idle.
It’s going to be an artifact of the data collection method. The sample size is extremely small and taken during the period where top has never stopped running.