add timing results on cc2.8xlarge (64 vCPU)

Parallel_Algorithm/MPI/test_hybrid.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
 
-np_list='1 2 4 8'
-thread_list='1 2'
+np_list='1 2 4 8 16 32 64'
+thread_list='1 2 4 8'
 
 for np in $np_list
 do

Parallel_Algorithm/MPI/test_multiprocess.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 
-np_list='1 2 4 8'
+np_list='1 2 4 8 16 32 64'
 
 export OMP_NUM_THREADS=1 # make sure OpenMP disabled 
 for np in $np_list

Parallel_Algorithm/OpenMP/test_multithread.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 
-thread_list='1 2 4 8'
+thread_list='1 2 4 8 16 32 64'
 
 for thread in $thread_list
 do

Timing_Results/log/Blobs_MPI.log

@@ -0,0 +1,147 @@
+
+=========================================
+=========================================
+testing with 1 processes
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+Last element in global array: -5.464932 
+Last element after scattering 0: -5.464932 
+=====Applying K-mean======
+Best inertia: 2308841.500000 
+I/O time use (ms): 3.746011 
+Kmean total time use (ms): 2327.527416 
+
+(sub-component timing not accurate) 
+E-step time use (ms): 2080.869320 
+M-step-1st-half time use (ms): 245.803846 
+M-step-2nd-half time use (ms): 0.497344 
+
+=========================================
+=========================================
+testing with 2 processes
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+Last element in global array: -5.464932 
+Last element after scattering 1: -5.464932 
+=====Applying K-mean======
+Best inertia: 2308842.000000 
+I/O time use (ms): 3.857213 
+Kmean total time use (ms): 1370.546596 
+
+(sub-component timing not accurate) 
+E-step time use (ms): 1208.027011 
+M-step-1st-half time use (ms): 161.206449 
+M-step-2nd-half time use (ms): 0.596461 
+
+=========================================
+=========================================
+testing with 4 processes
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+Last element in global array: -5.464932 
+Last element after scattering 3: -5.464932 
+=====Applying K-mean======
+Best inertia: 2308840.750000 
+I/O time use (ms): 4.113028 
+Kmean total time use (ms): 889.274717 
+
+(sub-component timing not accurate) 
+E-step time use (ms): 783.437196 
+M-step-1st-half time use (ms): 103.108949 
+M-step-2nd-half time use (ms): 0.884013 
+
+=========================================
+=========================================
+testing with 8 processes
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+Last element in global array: -5.464932 
+Last element after scattering 7: -5.464932 
+=====Applying K-mean======
+Best inertia: 2308833.500000 
+I/O time use (ms): 4.099924 
+Kmean total time use (ms): 490.620752 
+
+(sub-component timing not accurate) 
+E-step time use (ms): 421.552152 
+M-step-1st-half time use (ms): 63.779867 
+M-step-2nd-half time use (ms): 1.227340 
+
+=========================================
+=========================================
+testing with 16 processes
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+Last element in global array: -5.464932 
+Last element after scattering 15: -5.464932 
+=====Applying K-mean======
+Best inertia: 2308833.500000 
+I/O time use (ms): 4.935520 
+Kmean total time use (ms): 298.665461 
+
+(sub-component timing not accurate) 
+E-step time use (ms): 221.786738 
+M-step-1st-half time use (ms): 67.226194 
+M-step-2nd-half time use (ms): 1.788349 
+
+=========================================
+=========================================
+testing with 32 processes
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+Last element in global array: -5.464932 
+Last element after scattering 31: -5.464932 
+=====Applying K-mean======
+Best inertia: 2308833.000000 
+I/O time use (ms): 9.008225 
+Kmean total time use (ms): 280.026070 
+
+(sub-component timing not accurate) 
+E-step time use (ms): 183.410053 
+M-step-1st-half time use (ms): 76.902479 
+M-step-2nd-half time use (ms): 1.928765 
+
+=========================================
+=========================================
+testing with 64 processes
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+Last element in global array: -5.464932 
+Last element after scattering 63: -5.464932 
+=====Applying K-mean======
+Best inertia: 2308827.750000 
+I/O time use (ms): 19.194771 
+Kmean total time use (ms): 985.778767 
+
+(sub-component timing not accurate) 
+E-step time use (ms): 106.266378 
+M-step-1st-half time use (ms): 605.020996 
+M-step-2nd-half time use (ms): 2.200050 

Timing_Results/log/Blobs_OpenMP.log

@@ -0,0 +1,126 @@
+
+=========================================
+=========================================
+testing with 1 threads
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+=====Applying K-mean======
+=====writting data finished======
+Best inertia: 2308841.500000 
+Kmean total time use (ms): 2539.128065 
+E-step time use (ms): 2293.212175 
+M-step-1st-half time use (ms): 245.102882 
+M-step-2nd-half time use (ms): 0.499487 
+I/O time use (ms): 2.501011 
+
+=========================================
+=========================================
+testing with 2 threads
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+=====Applying K-mean======
+=====writting data finished======
+Best inertia: 2308842.000000 
+Kmean total time use (ms): 1731.789112 
+E-step time use (ms): 1426.280260 
+M-step-1st-half time use (ms): 304.410219 
+M-step-2nd-half time use (ms): 0.734806 
+I/O time use (ms): 1.926899 
+
+=========================================
+=========================================
+testing with 4 threads
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+=====Applying K-mean======
+=====writting data finished======
+Best inertia: 2308840.500000 
+Kmean total time use (ms): 1336.784840 
+E-step time use (ms): 934.716463 
+M-step-1st-half time use (ms): 399.653912 
+M-step-2nd-half time use (ms): 1.964569 
+I/O time use (ms): 2.602816 
+
+=========================================
+=========================================
+testing with 8 threads
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+=====Applying K-mean======
+=====writting data finished======
+Best inertia: 2308833.500000 
+Kmean total time use (ms): 900.095940 
+E-step time use (ms): 490.777731 
+M-step-1st-half time use (ms): 407.839060 
+M-step-2nd-half time use (ms): 1.050949 
+I/O time use (ms): 2.654076 
+
+=========================================
+=========================================
+testing with 16 threads
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+=====Applying K-mean======
+=====writting data finished======
+Best inertia: 2308838.000000 
+Kmean total time use (ms): 741.780996 
+E-step time use (ms): 309.553146 
+M-step-1st-half time use (ms): 430.645466 
+M-step-2nd-half time use (ms): 1.129389 
+I/O time use (ms): 2.640963 
+
+=========================================
+=========================================
+testing with 32 threads
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+=====Applying K-mean======
+=====writting data finished======
+Best inertia: 2308832.750000 
+Kmean total time use (ms): 782.614946 
+E-step time use (ms): 210.109472 
+M-step-1st-half time use (ms): 570.574045 
+M-step-2nd-half time use (ms): 1.244783 
+I/O time use (ms): 2.243996 
+
+=========================================
+=========================================
+testing with 64 threads
+reading data 
+Number of samples: 20000 
+Number of features: 30 
+Number of clusters: 8 
+Number of repeated runs: 20 
+=====reading data finished======
+=====Applying K-mean======
+=====writting data finished======
+Best inertia: 2308833.750000 
+Kmean total time use (ms): 1440.789938 
+E-step time use (ms): 1023.807049 
+M-step-1st-half time use (ms): 415.407658 
+M-step-2nd-half time use (ms): 1.029730 
+I/O time use (ms): 2.650023