FAQ  •  Login

HW 10 Timing Results

<<

chelseadangelo

Newbie
Newbie

Posts: 6

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 2:49 pm

Re: HW 10 Timing Results

Compiler: gcc
Processor: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz

Serial Time: 127.34 ms
OpenMP Time: 13.13 ms
Results match reference value to 1e-7.
<<

timhaines

Newbie
Newbie

Posts: 34

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 2:50 pm

Re: HW 10 Timing Results

  Code:
.--------------------------------------.
| AMD FX-8350 8-Core @ 4GHz (with AVX) |
+----------+---------------+-----------+
| threads  | Runtime (ms)  | % diff    |
+----------+---------------+-----------+
|        1 |       9.59619 | 1.287e-04 |
|        2 |       4.85985 | 1.287e-04 |
|        4 |       2.45891 | 1.287e-04 |
|        8 |       1.79685 | 1.287e-04 |
|       16 |       3.36288 | 1.287e-04 |
|       32 |       4.09105 | 1.287e-04 |
|       64 |       4.85315 | 1.287e-04 |
'----------+---------------+-----------'


  Code:
.---------------------------------------------.
| Intel Xeon E5520 16-core @ 2.27GHz (no AVX) |
+------------+-----------------+--------------+
| threads    | Runtime (ms)    | % diff       |
+------------+-----------------+--------------+
|          1 |           111.4 | 9.555e-03    |
|          2 |         59.1456 | 9.555e-03    |
|          4 |         29.8255 | 9.555e-03    |
|          8 |         17.3534 | 9.555e-03    |
|         16 |         11.0013 | 9.555e-03    |
|         32 |         12.2149 | 9.555e-03    |
|         64 |         12.4073 | 9.555e-03    |
'------------+-----------------+--------------'


The '% diff' column reports the percent difference between the calculated value and the reference value provided from MATLAB. The AVX version does not work on Euler because the Xeon E5520 only supports up to SSE3.
- Tim
<<

pkrishna

Newbie
Newbie

Posts: 9

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 2:51 pm

Re: HW 10 Timing Results

processor : 1
vendor_id : GenuineIntel
cpu family : 6
model : 26
model name : Intel(R) Xeon(R) CPU E5520 @ 2.27GHz
stepping : 5
microcode : 0x19
cpu MHz : 1600.000
cache size : 8192 KB
physical id : 0
siblings : 8
core id : 1
cpu cores : 4
apicid : 2
initial apicid : 2
fpu : yes
fpu_exception : yes
cpuid level : 11
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm dca sse4_1 sse4_2 popcnt lahf_lm ida dtherm tpr_shadow vnmi flexpriority ept vpid
bogomips : 4533.08
clflush size : 64

Serial time 128.85 ms
opeMp time 13.41 ms
no. of threads 16
<<

mojeskoli

Newbie
Newbie

Posts: 12

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 3:02 pm

Re: HW 10 Timing Results

For sequential (one threads) : ~173 milliseconds
For max config : ~ 20 milliseconds
<<

juanpablocruzbastida

Newbie
Newbie

Posts: 9

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 4:21 pm

Re: HW 10 Timing Results

Sequential implementation: I = 32.121040666359477
Execution time: 170.000000 ms
Parallel implementation: I = 30.243924624064814
Execution time: 0.066927 ms

AMD processor, g++ compiler
<<

jonathoncrandallmagana

Newbie
Newbie

Posts: 14

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 4:39 pm

Re: HW 10 Timing Results

- One computational thread (sequential execution)
I =32.121089 (Should be approx. 32.1210)
Time: 0.173255 ms

- Using OpenMP on an Intel processor (which lists 16 cores):
// 16 threads
I =32.121089 (Should be approx. 32.1210)
Time: 0.018790 ms
The processor lists all 16 cores as:
Intel(R) Xeon(R) CPU E5520 @ 2.27GHz
I compiled the code with the following 2 commands:
g++ -c hw10p1.c -o test.o -fopenmp
g++ test.o -o hw10p1 -fopenmp -lpthread
<<

deeptibhaskar

Newbie
Newbie

Posts: 11

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 5:02 pm

Re: HW 10 Timing Results

Timing results:

Sequential:
Value of result is: 32.121089
Elapsed time = 129.019ms
numThreads = 1

Open MP
Value of result is: 32.121089
Elapsed time = 18.555ms
numThreads = 16
<<

alvarolinares

Newbie
Newbie

Posts: 15

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 5:13 pm

Re: HW 10 Timing Results

Serial timing: 0.242s (1 CPU)
Parallel timing: 0.0208 s (16 CPU)
Same result within 1e-13.


Processor info:
Architecture: x86_64
Byte Order: Little Endian
CPU(s): 16
Thread(s) per core: 2
Core(s) per socket: 4
Socket(s): 2
NUMA node(s): 2
Model name: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz

Compiler:
gcc version 4.8.3 20140911 (Red Hat 4.8.3-9) (GCC)
<<

chandanahosamanekabbali

Newbie
Newbie

Posts: 9

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 5:19 pm

Re: HW 10 Timing Results

Time required for serial = 37.575101 ms
Time required for parallel omp = 6.674458 ms
Number of processors = 16
Max number of threads = 16

Compiler : gcc (GCC) 4.8.3 20140911 (Red Hat 4.8.3-9)
CPU info:
Architecture: x86_64
CPU(s): 16
On-line CPU(s) list: 0-15
Thread(s) per core: 2
Model name: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz
<<

jamessawicki

Newbie
Newbie

Posts: 15

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 7:56 pm

Re: HW 10 Timing Results

Compiler and optimization: g++ -O4

CPU: Intel Xeon E5520 @ 2.27 GHz

Single-threaded result: 32.121011
Single-threaded time (average): 122.066 ms
Number of threads: 16
Multi-threaded result: 32.121425
Multi-threaded time (average): 12.596 ms
Reference result: 32.121040688226245
<<

xiluwang

Newbie
Newbie

Posts: 5

Joined: Tue Sep 08, 2015 6:03 pm

Unread post Wed Nov 25, 2015 8:26 pm

Re: HW 10 Timing Results

The result of sequential integral is: 32.121041
The run time of sequential integral is: 176.741301 ms

The result of openMP integral is: 32.121041
The run time of openMP integral is: 15.783226 ms

The speed up by openMP is about 11.2X
<<

yingzhang

Newbie
Newbie

Posts: 7

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 9:09 pm

Re: HW 10 Timing Results

On euler node, compile with "-O3" Optimization.

Max number of threads: 16
Serial Result: 32.121041
Serial Time: 144.616738ms
Parallel Result: 32.121108
Parallel Time: 13.227656ms
<<

shashankgupta

Newbie
Newbie

Posts: 16

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 10:08 pm

Re: HW 10 Timing Results

Serial Compute
Answer = 32.111652
Time = 146.424043 ms
Parallel compute
Answer = 32.121052
Time = 17.519188 ms
Number of processors = 16
Max threads = 16
Speedup : 8.6x
<<

kaziahmed

Newbie
Newbie

Posts: 17

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 11:01 pm

Re: HW 10 Timing Results

Processor: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz
Sequential result: 32.12108893283
Sequential timing: 211.13998070 ms

Open MP with 16 threads
Parallel result: 32.12108893283
Parallel timing: 20.78284975 ms

Timing with omp_get_wtime()
<<

daynahashemi

Newbie
Newbie

Posts: 17

Joined: Fri Sep 04, 2015 12:51 pm

Unread post Wed Nov 25, 2015 11:49 pm

Re: HW 10 Timing Results

Compile information: gcc (GCC) 4.8.3 20140911 (Red Hat 4.8.3-9)
Number of threads = 16
Sequential Implementation Value for I: 32.1210407
Sequential execution time: 130 ms
Parallel Implementation Value for I: 32.1210407
Parallel execution time: 17.5431 ms
PreviousNext

Return to ME759 Fall 2015: High Performance Computing

Who is online

Users browsing this forum: No registered users and 1 guest

cron
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group.
Designed by ST Software.