Sometimes you want to trace a specific process, and strace does that nicely. But other times you may need to see more about the whole system as it reacts with your program. That's what LTT is for. As the docs explain:
That being said, these tools often only give a snapshot of the system. You are rarely able to know exactly what happens in a system at every moment in time. You can only try to know by polling faster, with the perverted effects of such a technique. LTT fills this gap by providing its user with all the information required to reconstruct a system's behavior during a certain period of time. One can know exactly the dynamics of a system. Why do certain synchronization problems occur? What exactly happens to an application when a packet is received for it? Overall, where do all the applications that I use spend their time? Where are the I/O latencies in a given application? etc.
Got something to add? Send me email.
More Articles by Tony Lawrence © 2011-03-20 Tony Lawrence
There are two major products that came out of Berkeley: LSD and UNIX. We do not believe this to be a coincidence. (Jeremy S. Anderson)