Lines Matching refs:as

11 In addition, as described in
16 such as monitoring application codes and tips for accurate profiling.
26   and require little overhead; thus, `-log_view` is intended as the
55 as discussed in the following section.
58 one *flop* as one operation of any of the following types:
69 of the PETSc library, as shown in
118 routine such as `KSPSolve()` include all of the operations accumulated
125 operations are performed when using GMRES and ILU, as described above),
136 We next discuss performance summaries for parallel programs, as shown
152 computational rates, and message-passing activity (such as the number
154 various user-defined stages of monitoring (as discussed in
156 various phases of computation then follow (as shown separately here in
213 computation, as given in the table within
215 phase presents the maximum times and flop rates over all processes, as
216 well as the ratio of maximum to minimum times and flop rates for all
218 a given phase are well balanced among the processes; as the ratio
227 *Note*: Total computational rates $<$ 1 MFlop are listed as 0 in
296 operations, as given by the routines `VecScatterBegin()` and
328 `logfile`. Events can be deactivated as described in
332 application as any MPI application, and follow the MPI implementation’s
343 either as an XML file or as a text file in a format suitable for viewing as
371 information by profiling their application codes as well. The basic
413 calls will be automatically timed and logged as part of the specified
487 considerable overhead of paging, as discussed in
502 The `event` may be either a predefined PETSc event (as listed in the
504 `PetscLogEventRegister()` (as described in
537 algorithms and tuning program performance. For example, as discussed in
557 seeing all lower level data as well (e.g., `Mat`).
572 Application programmers can log their own messages, as well, by using the
581 methods, we use a statement such as
599 returned in order to express fractions of a second. In addition, as
692 or Open MPI - we can run a parallel job on 4 MPI tasks as:
709 For AMD GPUs, log events registered to PETSc can be displayed as ranges in trace files generated by…