Lines Matching full:system
33 of the system.
36 System maintenance tools
39 observing the basic actions of the system.
47 when the system is active you can judge the system activity in several
50 Ideally, to have a balanced system in activity,
57 If the system is busy, then the count of active jobs may be large,
66 when the system is busy (a ``vmstat 5'' gives all the
67 numbers computed by the system), you can find
72 that are doing high-speed non-buffered input/output, then the system
74 It is often possible to pin down the cause of high system time by
76 activity (in) or system call activity (sy). Long term measurements
80 per second and an average of 90 system calls per second.
82 If the system is heavily loaded, or if you have little memory
83 for your load (1 megabyte is little in our environment), then the system
85 reduction in the system responsiveness and long pauses when interactive
94 It also gives the percentage of time the system has
96 in system mode, and idling.
111 to pinpoint bottlenecks. Under heavy system load, disk
115 attached to the same controller, the system will
124 output rates are system specific. Screen editors
137 The program counter sampling can be driven by the system clock,
139 The latter is highly recommended as use of the system clock results
143 Once a profiling system has been booted statistic gathering is
152 to obtain information regarding the system's operation.
155 and a dynamic call graph of the executing system.
166 A profiled system is about 5-10% larger in its text space because of
168 When the system executes,
174 The overhead for running a profiled system varies;
176 of the system time spent in the profiling code.
177 Thus the system is noticeably slower than an unprofiled system,
190 system is operating. Each trace record contains a time stamp
202 distributed system is calculated from the kernel's time of day
220 system load.
223 the file system buffer cache. The sampling program maintained a
231 system operations. Operations such as system call overhead and
233 overall performance of a system. Because of the drastic changes in
234 the system between 4.1BSD and 4.2BSD, it was important to verify
239 When we suspected a specific problem with the system,
250 in the long term system performance is negligible,
252 but the long term system performance improves significantly.