Skip to main content

Advances, Systems and Applications

Table 11 Accumulated Exclusive time of processes of MPIRandomAccess in case of 8 threads

From: Investigating performance metrics for container-based HPC environments using x86 and OpenPOWER systems

  

x86

OpenPOWER

Processes

Interpretation

Exclusive Execution Time(Sec)

% of Total Execution Time

Exclusive Execution Time(Sec)

% of Total Execution Time

Unknown

not reported

1.051

37.73

1.433

19.59

MPIDI_CH3I_Progress

maintain process progress state for asynchronous communications

0.554

19.90

1.032

14.10

MPID_nem_tcp_connpoll

tcp communication between processes

0.112

04.02

0.924

12.63

Tau_global_decr_insideTAU:TauCAPI.cpp

global decrement inside TAU

-

-

0.486

06.64

Tau_global_incr_insideTAU:TauCAPI.cpp

global increment inside TAU

0.026

00.93

0.317

04.33

HPCC_PoolGetObj

return pool of processes represented by buckets

0.311

11.17

-

-

HPCC_PoolReturnObj

return pool of processes represented by buckets

0.173

06.21

0.265

03.62

FunctionInfo::IsThrottled() const:FunctionInfo.h

throttled timers

-

-

0.251

03.43

HPCC_ra_Heap_IncrementKey

Random access heap key increment

-

-

0.223

03.05

MPID_nem_network_poll

network polling for communicating threads

-

-

0.203

02.77

Tau_start_timer:TauCAPI.cpp

start timer for a particular thread

0.200

02.73

0.200

02.73

Tau_global_getLightsOut:TauCAPI.cpp

to control disable profiling

0.194

02.66

0.194

02.65

Tau_lite_stop_timer:TauCAPI.cpp

stop timer for a phase

0.129

04.63

0.419

05.73

Tau_lite_stop_timer:TauCAPI.cpp

stop timer for a phase

0.079

02.84

0.184

02.51

HPCC_InsertUpdate

maintains the updates for each PE

-

-

0.180

02.46

Tau_get_thread:TauCAPI.cpp

Return id of the worker thread

-

-

0.172

02.35

MPID_nem_barrier

thread barrier for synchronization

-

-

0.157

02.15

Tau_memory_wrapper_disable:TauMemory.cpp

memory wrapper disable library

0.135

04.85

0.137

01.87

HPCC_GetUpdates

Each process (PE) maintains a set of destination PE

-

-

0.102

01.39

PMPI_Testany

used to wait for the completion of one out of several operations

0.074

02.77

-

-