Friday, 15 November 2013

Performance of new AWS c3.8xlarge instance

Benchmark results for new c3.8xlarge AWS instance:


photo 1


photo 2

Using username "ec2-user".
Authenticating with public key "imported-openssh-key"
Last login: Fri Nov 15 02:29:57 2013 from 129.94.41.203

       __|  __|_  )
       _|  (     /   Amazon Linux AMI
      ___|\___|___|

https://aws.amazon.com/amazon-linux-ami/2013.09-release-notes/
No packages needed for security; 4 packages available
Run "sudo yum update" to apply all updates.
[ec2-user@ip-10-250-2-252 ~]$
[ec2-user@ip-10-250-2-252 ~]$ cat /proc/cpuinfo | grep -i processor | wc -l
32
[ec2-user@ip-10-250-2-252 ~]$ cat /proc/meminfo
MemTotal:       61603084 kB
MemFree:        61214964 kB
Buffers:            6508 kB
Cached:           181536 kB
SwapCached:            0 kB
Active:           169776 kB
Inactive:          29000 kB
Active(anon):      10800 kB
Inactive(anon):        8 kB
Active(file):     158976 kB
Inactive(file):    28992 kB
Unevictable:           0 kB
Mlocked:               0 kB
SwapTotal:             0 kB
SwapFree:              0 kB
Dirty:                 4 kB
Writeback:             0 kB
AnonPages:         10788 kB
Mapped:             6480 kB
Shmem:                48 kB
Slab:              50124 kB
SReclaimable:      20456 kB
SUnreclaim:        29668 kB
KernelStack:        1712 kB
PageTables:         1900 kB
NFS_Unstable:          0 kB
Bounce:                0 kB
WritebackTmp:          0 kB
CommitLimit:    30801540 kB
Committed_AS:      53016 kB
VmallocTotal:   34359738367 kB
VmallocUsed:      114088 kB
VmallocChunk:   34359624263 kB
AnonHugePages:         0 kB
HugePages_Total:       0
HugePages_Free:        0
HugePages_Rsvd:        0
HugePages_Surp:        0
Hugepagesize:       2048 kB
DirectMap4k:    62922752 kB
DirectMap2M:           0 kB
[ec2-user@ip-10-250-2-252 ~]$
[ec2-user@ip-10-250-2-252 Geekbench-2.4.0-Linux]$ df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/xvda1            7.9G 1000M  6.9G  13% /
tmpfs                  30G     0   30G   0% /dev/shm
[ec2-user@ip-10-250-2-252 Geekbench-2.4.0-Linux]$ dd if=/dev/zero of=bigfile bs=1024k count=5000
5000+0 records in
5000+0 records out
5242880000 bytes (5.2 GB) copied, 123.612 s, 42.4 MB/s

<....>

[ec2-user@ip-10-250-2-252 Geekbench-2.4.0-Linux]$


[ec2-user@ip-10-250-2-252 Geekbench-2.4.0-Linux]$ ./geekbench_x86_64
Geekbench 2.4.0 : http://www.primatelabs.com/geekbench/

System Information
  Operating System      Linux 3.4.68-59.97.amzn1.x86_64 x86_64
  Model
  Motherboard
  Processor                   Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz @ 2.80 GHz
                        1 Processor, 32 Threads
  Processor ID          GenuineIntel Family 6 Model 62 Stepping 4
  L1 Instruction Cache  32.0 KB x 16
  L1 Data Cache         32.0 KB x 16
  L2 Cache              256 KB x 16
  L3 Cache              25.0 MB
  Memory                58.7 GB
  BIOS

Integer
  Blowfish
    single-threaded scalar   2121 ||||||||
    multi-threaded scalar   63260 ||||||||||||||||||||||||||||||||||||||||
  Text Compress
    single-threaded scalar   3027 ||||||||||||
    multi-threaded scalar   61756 ||||||||||||||||||||||||||||||||||||||||
  Text Decompress
    single-threaded scalar   3683 ||||||||||||||
    multi-threaded scalar   77284 ||||||||||||||||||||||||||||||||||||||||
  Image Compress
    single-threaded scalar   2712 ||||||||||
    multi-threaded scalar   58859 ||||||||||||||||||||||||||||||||||||||||
  Image Decompress
    single-threaded scalar   3112 ||||||||||||
    multi-threaded scalar   58788 ||||||||||||||||||||||||||||||||||||||||
  Lua
    single-threaded scalar   5294 |||||||||||||||||||||
    multi-threaded scalar   99794 ||||||||||||||||||||||||||||||||||||||||

Floating Point
  Mandelbrot
    single-threaded scalar   2740 ||||||||||
    multi-threaded scalar   87872 ||||||||||||||||||||||||||||||||||||||||
  Dot Product
    single-threaded scalar   4272 |||||||||||||||||
    multi-threaded scalar  108177 ||||||||||||||||||||||||||||||||||||||||
    single-threaded vector   6305 |||||||||||||||||||||||||
    multi-threaded vector  123948 ||||||||||||||||||||||||||||||||||||||||
  LU Decomposition
    single-threaded scalar   3302 |||||||||||||
    multi-threaded scalar   22878 ||||||||||||||||||||||||||||||||||||||||
  Primality Test
    single-threaded scalar   6271 |||||||||||||||||||||||||
    multi-threaded scalar  101071 ||||||||||||||||||||||||||||||||||||||||
  Sharpen Image
    single-threaded scalar   6661 ||||||||||||||||||||||||||
    multi-threaded scalar  175624 ||||||||||||||||||||||||||||||||||||||||
  Blur Image
    single-threaded scalar   2780 |||||||||||
    multi-threaded scalar   78959 ||||||||||||||||||||||||||||||||||||||||

Memory
  Read Sequential
    single-threaded scalar   5639 ||||||||||||||||||||||
  Write Sequential
    single-threaded scalar   7619 ||||||||||||||||||||||||||||||
  Stdlib Allocate
    single-threaded scalar   5306 |||||||||||||||||||||
  Stdlib Write
    single-threaded scalar   1381 |||||
  Stdlib Copy
    single-threaded scalar   1808 |||||||

Stream
  Stream Copy
    single-threaded scalar   6149 ||||||||||||||||||||||||
    single-threaded vector   6257 |||||||||||||||||||||||||
  Stream Scale
    single-threaded scalar   6547 ||||||||||||||||||||||||||
    single-threaded vector   6121 ||||||||||||||||||||||||
  Stream Add
    single-threaded scalar   5759 |||||||||||||||||||||||
    single-threaded vector   5873 |||||||||||||||||||||||
  Stream Triad
    single-threaded scalar   6262 |||||||||||||||||||||||||
    single-threaded vector   4392 |||||||||||||||||

Benchmark Summary
  Integer Score             36640 ||||||||||||||||||||||||||||||||||||||||
  Floating Point Score      52204 ||||||||||||||||||||||||||||||||||||||||
  Memory Score               4350 |||||||||||||||||
  Stream Score               5920 |||||||||||||||||||||||

  Geekbench Score           32557 ||||||||||||||||||||||||||||||||||||||||

Upload results to the Geekbench Browser? [Y/n]y

Uploading results to the Geekbench Browser. This could take a minute or two
depending on the speed of your internet connection.

Upload succeeded. Visit the following link and view your results online:

[ec2-user@ip-10-250-2-252 Geekbench-2.4.0-Linux]$


<….>

[ec2-user@ip-10-250-2-252 Geekbench-3.1.2-Linux]$ ./geekbench_x86_64
Geekbench 3.1.2 : http://www.primatelabs.com/geekbench/

System Information
  Operating System      Linux 3.4.68-59.97.amzn1.x86_64 x86_64
  Model                 N/A
  Motherboard           N/A
  Processor                   Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz @ 2.80 GHz
                        1 Processor, 32 Threads
  Processor ID          GenuineIntel Family 6 Model 62 Stepping 4
  L1 Instruction Cache  32.0 KB x 16
  L1 Data Cache         32.0 KB x 16
  L2 Cache              256 KB x 16
  L3 Cache              25.0 MB
  Memory                58.7 GB
  BIOS
  Compiler              Clang 3.3 (tags/RELEASE_33/final)

Integer
  AES
    single-core        2536          2.17 GB/sec
    multi-core        38531          33.0 GB/sec
  Twofish
    single-core        2716         152.4 MB/sec
    multi-core        61321          3.36 GB/sec
  SHA1
    single-core        3151         342.1 MB/sec
    multi-core        51862          5.50 GB/sec
  SHA2
    single-core        3433         148.6 MB/sec
    multi-core        52594          2.22 GB/sec
  BZip2 Compress
    single-core        2532          10.3 MB/sec
    multi-core        52007         211.4 MB/sec
  BZip2 Decompress
    single-core        2530          13.7 MB/sec
    multi-core        52614         285.2 MB/sec
  JPEG Compress
    single-core        2793     38.9 Mpixels/sec
    multi-core        59007    822.0 Mpixels/sec
  JPEG Decompress
    single-core        3441     85.1 Mpixels/sec
    multi-core        61620     1.52 Gpixels/sec
  PNG Compress
    single-core        2692     2.15 Mpixels/sec
    multi-core        58812     47.0 Mpixels/sec
  PNG Decompress
    single-core        2730     31.5 Mpixels/sec
    multi-core        58539    674.9 Mpixels/sec
  Sobel
    single-core        3454    125.7 Mpixels/sec
    multi-core        58656     2.13 Gpixels/sec
  Lua
    single-core        2934          2.64 MB/sec
    multi-core        55985          50.3 MB/sec
  Dijkstra
    single-core        2284      8.20 Mpairs/sec
    multi-core        30803     110.5 Mpairs/sec

Floating Point
  BlackScholes
    single-core        2349      10.5 Mnodes/sec
    multi-core        49401     219.8 Mnodes/sec
  Mandelbrot
    single-core        2686          2.75 Gflops
    multi-core        71811          73.6 Gflops
  Sharpen Filter
    single-core        2323          1.72 Gflops
    multi-core        47351          35.1 Gflops
  Blur Filter
    single-core        1929          1.84 Gflops
    multi-core        44703          42.6 Gflops
  SGEMM
    single-core        3472          9.73 Gflops
    multi-core        58402         163.6 Gflops
  DGEMM
    single-core        3390          4.98 Gflops
    multi-core        58441          85.9 Gflops
  SFFT
    single-core        2758          2.91 Gflops
    multi-core        43973          46.4 Gflops
  DFFT
    single-core        2909          2.65 Gflops
    multi-core        50320          45.8 Gflops
  N-Body
    single-core        4548      1.69 Mpairs/sec
    multi-core        77548      28.8 Mpairs/sec
  Ray Trace
    single-core        3625     4.27 Mpixels/sec
    multi-core        69037     81.4 Mpixels/sec

Memory
  Stream Copy
    single-core        1156          4.61 GB/sec
    multi-core         2284          9.11 GB/sec
  Stream Scale
    single-core        2296          9.17 GB/sec
    multi-core         4585          18.3 GB/sec
  Stream Add
    single-core        2189          9.90 GB/sec
    multi-core         4184          18.9 GB/sec
  Stream Triad
    single-core        2226          9.78 GB/sec
    multi-core         4265          18.7 GB/sec

Benchmark Summary
  Integer Score              2839  52377
  Floating Point Score       2913  56005
  Memory Score               1896   3697

  Geekbench Score            2680  44092

Upload results to the Geekbench Browser? [Y/n]y

Uploading results to the Geekbench Browser. This could take a minute or two
depending on the speed of your internet connection.

Upload succeeded. Visit the following link and view your results online:


<…>


[ec2-user@ip-10-250-2-252 x86_64-linux-gnu]$ ./mhz
3101 MHz, 0.3225 nanosec clock
[ec2-user@ip-10-250-2-252 x86_64-linux-gnu]$ ./stream
STREAM copy latency: 0.96 nanoseconds
STREAM copy bandwidth: 16751.00 MB/sec
STREAM scale latency: 0.97 nanoseconds
STREAM scale bandwidth: 16446.80 MB/sec
STREAM add latency: 1.30 nanoseconds
STREAM add bandwidth: 18450.18 MB/sec
STREAM triad latency: 1.29 nanoseconds
STREAM triad bandwidth: 18564.36 MB/sec
[ec2-user@ip-10-250-2-252 x86_64-linux-gnu]$ ./stream -P 10
STREAM copy latency: 3.32 nanoseconds
STREAM copy bandwidth: 48156.42 MB/sec
STREAM scale latency: 3.32 nanoseconds
STREAM scale bandwidth: 48241.43 MB/sec
STREAM add latency: 4.33 nanoseconds
STREAM add bandwidth: 55369.63 MB/sec
STREAM triad latency: 4.97 nanoseconds
STREAM triad bandwidth: 48317.76 MB/sec
[ec2-user@ip-10-250-2-252 x86_64-linux-gnu]$ ./lat_mem rd 512
-bash: ./lat_mem: No such file or directory
[ec2-user@ip-10-250-2-252 x86_64-linux-gnu]$ ./lat_mem_rd 512
"stride=64
0.00049 1.292
0.00098 1.292
0.00195 1.293
0.00293 1.293
0.00391 1.292
0.00586 1.294
0.00781 1.292
0.01172 1.293
0.01562 1.292
0.02344 1.293
0.03125 1.334
0.04688 3.845
0.06250 3.842
0.09375 3.846
0.12500 3.853
0.18750 4.016
0.25000 4.092
0.37500 4.172
0.50000 4.193
0.75000 4.221
1.00000 4.203
1.50000 4.220
2.00000 4.186
3.00000 4.258
4.00000 4.244
6.00000 4.270
8.00000 4.234
12.00000 4.263
16.00000 4.479
24.00000 6.715
32.00000 8.502
48.00000 8.977
64.00000 9.064
96.00000 9.166
128.00000 9.195
192.00000 9.299
256.00000 9.302
384.00000 9.326
512.00000 9.297
[ec2-user@ip-10-250-2-252 x86_64-linux-gnu]$



No comments:

Post a Comment