AWS or GCP?

Zooker

Zooker

Regular Member
Joined
Jan 13, 2010
Messages
358
Reaction score
120
I'm getting ready to move a site I've had in development up to the cloud so if this site takes off I am scale vertically or horizontally easier. I'm looking at both Amazon and Google for their cloud hosting abilities.

I've used both of them for CDN abilities, but not yet for hosting. I'm really looking for a good reason to choose one over the other (and I expect they both are fairly comparable best I can tell by reading).

Any thoughts good or bad?
 
Advertise on BHW
Are you interested in checking other options? I assume there some more CDN providers with horizontal and vertical scaling.
 
I've been doing a bit of research and I'm still not quite done. My plan is to get reasonable cloud hosting so that I can start small and scale vertically if I need to (as in, it takes off quickly) and then the bigger plan would be to scale horizontally.

I'm working on a spreadsheet that compares on-demand vs commitment for aws and gcp. comparing just the virtual cpu count and amount of ram, Google seems to be slightly in the lead by a few dollars. Couple that with the starting $300 credit you get there and it's looking tempting. That is a definite plus when you are starting out a new idea that the first 4-5 months of hosting are free in essence.

What I don't know yet is how the cpu cores compare (which depends largely on the host system) since in aws you have things like t4g and m6g and in gcp you have things like e2, n1, and n2. These are all general virtual machines so they're balances and fine when you have a site and database all in one.

So far, I haven't seen anything on either that gives it a clear advantage. aws does have the SES that lets you mail the first 32,000 emails through their mail server so less chance it'll bounce for normal-type emails like registration emails. it's possible on either system the IP you have on your virtual machine was blacklisted due to previous activity. GCP seems to recommend using Google Workspace (naturally), SendGrid, mailgun, and Mailjet as a way to send emails that are less likely to hit an issue.

Of course, since these are emails in relation to users registering so a server at either place might be fine just sending out email directly from the server. AWS does allow up to 62,000 emails using their service for free if you setup using aws ec2.

You can use AWS SES from any server anywhere it appears:
View attachment 176287

So if you're at say GCP and you send 5000 emails for the month, it's only 50 cents. Not enough in itself to affect server choice, but GCP has some weirdness it looks like with blocked ports and such around smtp and email. GCP specifically blocks outgoing port 25 which means they force you to use email sending capabilities outside of your cloud host. This I need to read more on.

Hope I'm not boring you guys with some of these details I am coming across. Some of these things you don't immediately see when you start looking at cloud hosting vs a traditional vps/dedicated. I'm going to compare the servers to compare the computing power and I'll add that spreadsheet on here in the next couple of days.
 
What reasons are you choosing AWS over GCP? Was it other services they offer at the same time, or was it footprint, or was it ease of use and upgrading and putting pieces together? I'm genuinely interested since I'm trying to evaluate them and ideally not change in the future and just expand out as necessary for capacity.

If anyone has any other thoughts around other cloud setups like Azure, etc, or even some hosting providers may have better offers and none of those clouds are really the best answer. Ideally we're talking about starting smaller and expanding out without being too painful.

AWS has 77 availability zones and 24 geographic regions and 210 network edge locations (so AWS likely has the latency edge)
GCP has 73 availability zones and 24 geographic regions and 144 network edge locations

AWS has 32% of cloud market share
GCP has 7% of the cloud market share.

I do see that a lot of tech companies are on AWS and there must be some reason behind that.

I have noticed from using both AWS and GCP for CDN purposes that the Google interface was a little tougher to do what you wanted to, and I'm a software engineer. It's like Linux vs Windows. The S3 interface was a little more intuitive. Maybe Amazon is just far more mature.

Are those some of the reasons you guys choose AWS? I haven't done cloud hosting before so I'm trying to come up to speed. Up to this point I've been a more classic VPS or dedicated host type person.
 
Last edited:
I use GCP for all my projects, very easy to setup and use especially if you code in google-owned products like Angular. I haven't used AWS so cant recommend which one is better!
 
I prefer GCP.
It's easier to manage with the organizations, folders, and projects.
Altough AWS is cheaper for VM instances. Also for free / trial option AWS has more to offer.
Support, business needs, security, etc depend on you.
 
I've been doing some benchmarking and such. It appears with the latest Graviton2 aws cpus on the ARM build of Amazon Linux 2, that they have a distinct advantage. Comparing to digital ocean (I haven't run the unix benchmark on GCP yet), the same # of vcpu/amount of ram, the AWS m-series (m6g) seems to have a noticeable lead on the intel-based cpus.

I ran these last night
M6g.medium - 1 vCPU / 4GB ram
Code: 
========================================================================
  BYTE UNIX Benchmarks (Version 5.1.3)

   System: us-east-2.compute.internal: GNU/Linux
   OS: GNU/Linux -- 4.14.232-176.381.amzn2.aarch64 -- #1 SMP Wed May 19 00:32:11 UTC 2021
   Machine: aarch64 (aarch64)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   02:04:25 up 8 min,  1 user,  load average: 0.05, 0.13, 0.08; runlevel 2021-06-23

------------------------------------------------------------------------
Benchmark Run: Wed Jun 23 2021 02:04:25 - 02:32:21
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       41193382.1 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     5928.9 MWIPS (9.6 s, 7 samples)
Execl Throughput                               6991.3 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1032125.2 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          289077.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       2945000.1 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1781893.6 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 356317.4 lps   (10.0 s, 7 samples)
Process Creation                              15739.3 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   7461.0 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1011.6 lpm   (60.0 s, 2 samples)
System Call Overhead                        1704857.3 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   41193382.1   3529.9
Double-Precision Whetstone                       55.0       5928.9   1078.0
Execl Throughput                                 43.0       6991.3   1625.9
File Copy 1024 bufsize 2000 maxblocks          3960.0    1032125.2   2606.4
File Copy 256 bufsize 500 maxblocks            1655.0     289077.5   1746.7
File Copy 4096 bufsize 8000 maxblocks          5800.0    2945000.1   5077.6
Pipe Throughput                               12440.0    1781893.6   1432.4
Pipe-based Context Switching                   4000.0     356317.4    890.8
Process Creation                                126.0      15739.3   1249.2
Shell Scripts (1 concurrent)                     42.4       7461.0   1759.7
Shell Scripts (8 concurrent)                      6.0       1011.6   1685.9
System Call Overhead                          15000.0    1704857.3   1136.6
                                                                   ========
System Benchmarks Index Score                                        1739.7

M6g.large - 2 vCPU / 8GB ram
Code: 
========================================================================
   BYTE UNIX Benchmarks (Version 5.1.3)

   System: us-east-2.compute.internal: GNU/Linux
   OS: GNU/Linux -- 4.14.232-176.381.amzn2.aarch64 -- #1 SMP Wed May 19 00:32:11 UTC 2021
   Machine: aarch64 (aarch64)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   03:45:27 up 6 min,  1 user,  load average: 0.00, 0.11, 0.07; runlevel 2021-06-23

------------------------------------------------------------------------
Benchmark Run: Wed Jun 23 2021 03:45:27 - 04:13:23
2 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       41263290.3 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     5927.0 MWIPS (9.6 s, 7 samples)
Execl Throughput                               6736.6 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1032921.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          289492.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       2889370.4 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1780979.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 133135.4 lps   (10.0 s, 7 samples)
Process Creation                              11269.0 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   8678.3 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1667.9 lpm   (60.0 s, 2 samples)
System Call Overhead                        1729070.6 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   41263290.3   3535.8
Double-Precision Whetstone                       55.0       5927.0   1077.6
Execl Throughput                                 43.0       6736.6   1566.6
File Copy 1024 bufsize 2000 maxblocks          3960.0    1032921.0   2608.4
File Copy 256 bufsize 500 maxblocks            1655.0     289492.5   1749.2
File Copy 4096 bufsize 8000 maxblocks          5800.0    2889370.4   4981.7
Pipe Throughput                               12440.0    1780979.8   1431.7
Pipe-based Context Switching                   4000.0     133135.4    332.8
Process Creation                                126.0      11269.0    894.4
Shell Scripts (1 concurrent)                     42.4       8678.3   2046.8
Shell Scripts (8 concurrent)                      6.0       1667.9   2779.8
System Call Overhead                          15000.0    1729070.6   1152.7
                                                                   ========
System Benchmarks Index Score                                        1640.3

------------------------------------------------------------------------
Benchmark Run: Wed Jun 23 2021 04:13:23 - 04:41:19
2 CPUs in system; running 2 parallel copies of tests

Dhrystone 2 using register variables       82401974.2 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                    11853.4 MWIPS (9.6 s, 7 samples)
Execl Throughput                              10920.5 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1271002.9 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          367715.0 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       3121952.7 KBps  (30.0 s, 2 samples)
Pipe Throughput                             3543921.4 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 626170.7 lps   (10.0 s, 7 samples)
Process Creation                              22902.2 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  12668.6 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1703.3 lpm   (60.0 s, 2 samples)
System Call Overhead                        2726288.7 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   82401974.2   7061.0
Double-Precision Whetstone                       55.0      11853.4   2155.2
Execl Throughput                                 43.0      10920.5   2539.6
File Copy 1024 bufsize 2000 maxblocks          3960.0    1271002.9   3209.6
File Copy 256 bufsize 500 maxblocks            1655.0     367715.0   2221.8
File Copy 4096 bufsize 8000 maxblocks          5800.0    3121952.7   5382.7
Pipe Throughput                               12440.0    3543921.4   2848.8
Pipe-based Context Switching                   4000.0     626170.7   1565.4
Process Creation                                126.0      22902.2   1817.6
Shell Scripts (1 concurrent)                     42.4      12668.6   2987.9
Shell Scripts (8 concurrent)                      6.0       1703.3   2838.8
System Call Overhead                          15000.0    2726288.7   1817.5
                                                                   ========
System Benchmarks Index Score                                        2747.9

According to the runs I found over at https://vpscritic.com/p/digitalocean/ , the digital ocean ones seem to come in around maybe 1200 on a single cpu and 1650 on 2 cpus which is expected for the Intel CPUs. GCP uses similar CPUs so I expect their results to be in this ballpark.

I know comparing them to digital ocean is a little off topic, but it appears they are used often because they have the distinct advantage of having 4TB+ included bandwidth. So if you are hosting say a video site, I think i would avoid aws and gcp and go with digital ocean just for the money savings on bandwidth alone.

I may run some benchmarks on GCP tonight to compare. I do realize that it's just one metric and it needs to be setup as a web server and load tested. There are just so many options I am trying to find a way to comparing them concretely.
 
Got some more benchmarks for GCP. The n2d seems to be fairly comparable to the AWS m6g in terms of raw computing power. The m6g.large seems to be roughly 14% more powerful than the n2d-standard-2. The e2 and n2 series is most of the price of the n2d and far less power.

e2-standard-2 - 2 vCPU / 8GB ram
Code: 
========================================================================
   BYTE UNIX Benchmarks (Version 5.1.3)

   System: e2-standard-2: GNU/Linux
   OS: GNU/Linux -- 4.19.0-16-cloud-amd64 -- #1 SMP Debian 4.19.181-1 (2021-03-19)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
   CPU 0: Intel(R) Xeon(R) CPU @ 2.30GHz (4600.0 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   CPU 1: Intel(R) Xeon(R) CPU @ 2.30GHz (4600.0 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   02:07:23 up 3 min,  1 user,  load average: 0.34, 0.21, 0.08; runlevel Jun

------------------------------------------------------------------------
Benchmark Run: Thu Jun 24 2021 02:07:23 - 02:35:18
2 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       35374737.7 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     5851.8 MWIPS (9.0 s, 7 samples)
Execl Throughput                               3548.3 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        484624.8 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          136260.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1372681.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                              716275.0 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  93434.1 lps   (10.0 s, 7 samples)
Process Creation                               5349.6 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   8915.8 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1404.7 lpm   (60.0 s, 2 samples)
System Call Overhead                         425646.6 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   35374737.7   3031.3
Double-Precision Whetstone                       55.0       5851.8   1064.0
Execl Throughput                                 43.0       3548.3    825.2
File Copy 1024 bufsize 2000 maxblocks          3960.0     484624.8   1223.8
File Copy 256 bufsize 500 maxblocks            1655.0     136260.5    823.3
File Copy 4096 bufsize 8000 maxblocks          5800.0    1372681.5   2366.7
Pipe Throughput                               12440.0     716275.0    575.8
Pipe-based Context Switching                   4000.0      93434.1    233.6
Process Creation                                126.0       5349.6    424.6
Shell Scripts (1 concurrent)                     42.4       8915.8   2102.8
Shell Scripts (8 concurrent)                      6.0       1404.7   2341.2
System Call Overhead                          15000.0     425646.6    283.8
                                                                   ========
System Benchmarks Index Score                                         944.9

------------------------------------------------------------------------
Benchmark Run: Thu Jun 24 2021 02:35:18 - 03:03:15
2 CPUs in system; running 2 parallel copies of tests

Dhrystone 2 using register variables       50692476.6 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                    10556.2 MWIPS (9.1 s, 7 samples)
Execl Throughput                               4937.9 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        671261.6 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          181242.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1867892.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                              958844.6 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 235300.1 lps   (10.0 s, 7 samples)
Process Creation                              12577.6 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  10229.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1424.3 lpm   (60.0 s, 2 samples)
System Call Overhead                         565687.3 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   50692476.6   4343.8
Double-Precision Whetstone                       55.0      10556.2   1919.3
Execl Throughput                                 43.0       4937.9   1148.4
File Copy 1024 bufsize 2000 maxblocks          3960.0     671261.6   1695.1
File Copy 256 bufsize 500 maxblocks            1655.0     181242.3   1095.1
File Copy 4096 bufsize 8000 maxblocks          5800.0    1867892.9   3220.5
Pipe Throughput                               12440.0     958844.6    770.8
Pipe-based Context Switching                   4000.0     235300.1    588.3
Process Creation                                126.0      12577.6    998.2
Shell Scripts (1 concurrent)                     42.4      10229.9   2412.7
Shell Scripts (8 concurrent)                      6.0       1424.3   2373.8
System Call Overhead                          15000.0     565687.3    377.1
                                                                   ========
System Benchmarks Index Score                                        1398.5

N2d-standard-2 - 2 vCPU/8GB​

Code: 
========================================================================
   BYTE UNIX Benchmarks (Version 5.1.3)

   System: n2d-standard-2: GNU/Linux
   OS: GNU/Linux -- 4.19.0-16-cloud-amd64 -- #1 SMP Debian 4.19.181-1 (2021-03-19)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
   CPU 0: AMD EPYC 7B12 (4500.0 bogomips)
          Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   CPU 1: AMD EPYC 7B12 (4500.0 bogomips)
          Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   22:45:46 up 6 min,  1 user,  load average: 0.24, 0.09, 0.02; runlevel Jun

------------------------------------------------------------------------
Benchmark Run: Wed Jun 23 2021 22:45:46 - 23:13:42
2 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       39394690.3 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     5188.4 MWIPS (9.4 s, 7 samples)
Execl Throughput                               5953.4 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1178798.1 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          322358.2 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       3083983.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1925318.1 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  41618.5 lps   (10.0 s, 7 samples)
Process Creation                               9140.1 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  11464.0 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1963.5 lpm   (60.0 s, 2 samples)
System Call Overhead                        2564160.6 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   39394690.3   3375.7
Double-Precision Whetstone                       55.0       5188.4    943.3
Execl Throughput                                 43.0       5953.4   1384.5
File Copy 1024 bufsize 2000 maxblocks          3960.0    1178798.1   2976.8
File Copy 256 bufsize 500 maxblocks            1655.0     322358.2   1947.8
File Copy 4096 bufsize 8000 maxblocks          5800.0    3083983.0   5317.2
Pipe Throughput                               12440.0    1925318.1   1547.7
Pipe-based Context Switching                   4000.0      41618.5    104.0
Process Creation                                126.0       9140.1    725.4
Shell Scripts (1 concurrent)                     42.4      11464.0   2703.8
Shell Scripts (8 concurrent)                      6.0       1963.5   3272.5
System Call Overhead                          15000.0    2564160.6   1709.4
                                                                   ========
System Benchmarks Index Score                                        1579.0

------------------------------------------------------------------------
Benchmark Run: Wed Jun 23 2021 23:13:42 - 23:41:39
2 CPUs in system; running 2 parallel copies of tests

Dhrystone 2 using register variables       56285959.1 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     9488.6 MWIPS (9.4 s, 7 samples)
Execl Throughput                               8268.4 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1613417.0 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          438448.4 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       4471113.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                             2776636.2 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 444542.4 lps   (10.0 s, 7 samples)
Process Creation                              19715.4 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  14807.6 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   2171.7 lpm   (60.0 s, 2 samples)
System Call Overhead                        3960227.5 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   56285959.1   4823.1
Double-Precision Whetstone                       55.0       9488.6   1725.2
Execl Throughput                                 43.0       8268.4   1922.9
File Copy 1024 bufsize 2000 maxblocks          3960.0    1613417.0   4074.3
File Copy 256 bufsize 500 maxblocks            1655.0     438448.4   2649.2
File Copy 4096 bufsize 8000 maxblocks          5800.0    4471113.5   7708.8
Pipe Throughput                               12440.0    2776636.2   2232.0
Pipe-based Context Switching                   4000.0     444542.4   1111.4
Process Creation                                126.0      19715.4   1564.7
Shell Scripts (1 concurrent)                     42.4      14807.6   3492.4
Shell Scripts (8 concurrent)                      6.0       2171.7   3619.5
System Call Overhead                          15000.0    3960227.5   2640.2
                                                                   ========
System Benchmarks Index Score                                        2731.4

N2-standard-2 - 2vCPU / 8GB​

Code: 
========================================================================
   BYTE UNIX Benchmarks (Version 5.1.3)

   System: n2-standard-2: GNU/Linux
   OS: GNU/Linux -- 4.19.0-16-cloud-amd64 -- #1 SMP Debian 4.19.181-1 (2021-03-19)
   Machine: x86_64 (unknown)
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
   CPU 0: Intel(R) Xeon(R) CPU (5600.6 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   CPU 1: Intel(R) Xeon(R) CPU (5600.6 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   03:25:25 up  1:08,  2 users,  load average: 0.00, 0.98, 1.96; runlevel Jun

------------------------------------------------------------------------
Benchmark Run: Thu Jun 24 2021 03:25:25 - 03:53:16
2 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       43482898.8 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     7472.0 MWIPS (8.9 s, 7 samples)
Execl Throughput                               4661.7 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        866959.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          236192.6 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       2564770.2 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1312580.5 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  74241.5 lps   (10.0 s, 7 samples)
Process Creation                               8134.0 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  11182.4 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1789.3 lpm   (60.0 s, 2 samples)
System Call Overhead                         913846.6 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   43482898.8   3726.0
Double-Precision Whetstone                       55.0       7472.0   1358.5
Execl Throughput                                 43.0       4661.7   1084.1
File Copy 1024 bufsize 2000 maxblocks          3960.0     866959.5   2189.3
File Copy 256 bufsize 500 maxblocks            1655.0     236192.6   1427.1
File Copy 4096 bufsize 8000 maxblocks          5800.0    2564770.2   4422.0
Pipe Throughput                               12440.0    1312580.5   1055.1
Pipe-based Context Switching                   4000.0      74241.5    185.6
Process Creation                                126.0       8134.0    645.6
Shell Scripts (1 concurrent)                     42.4      11182.4   2637.4
Shell Scripts (8 concurrent)                      6.0       1789.3   2982.1
System Call Overhead                          15000.0     913846.6    609.2
                                                                   ========
System Benchmarks Index Score                                        1375.6

------------------------------------------------------------------------
Benchmark Run: Thu Jun 24 2021 03:53:16 - 04:21:09
2 CPUs in system; running 2 parallel copies of tests

Dhrystone 2 using register variables       57779721.4 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                    13102.7 MWIPS (9.0 s, 7 samples)
Execl Throughput                               6258.7 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        999519.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          262548.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       3008848.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1431339.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 338871.2 lps   (10.0 s, 7 samples)
Process Creation                              16042.0 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  13522.7 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1893.7 lpm   (60.0 s, 2 samples)
System Call Overhead                         960089.6 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   57779721.4   4951.1
Double-Precision Whetstone                       55.0      13102.7   2382.3
Execl Throughput                                 43.0       6258.7   1455.5
File Copy 1024 bufsize 2000 maxblocks          3960.0     999519.5   2524.0
File Copy 256 bufsize 500 maxblocks            1655.0     262548.7   1586.4
File Copy 4096 bufsize 8000 maxblocks          5800.0    3008848.9   5187.7
Pipe Throughput                               12440.0    1431339.8   1150.6
Pipe-based Context Switching                   4000.0     338871.2    847.2
Process Creation                                126.0      16042.0   1273.2
Shell Scripts (1 concurrent)                     42.4      13522.7   3189.3
Shell Scripts (8 concurrent)                      6.0       1893.7   3156.2
System Call Overhead                          15000.0     960089.6    640.1
                                                                   ========
System Benchmarks Index Score                                        1940.5
 
Last edited:
Something just happened with AWS that is weird enough I'm going to mention it to everyone.

Amazon moved my instance to a different region so in my console I suddenly couldn't find it when I logged in. I finally thought let's change the region view in the dashboard and I eventually found where it was. However I didn't move it ( I don't think you could anyways ) and I got no notification about the change. The IP didn't change and it's still in a location with the same base cost and it's right now in my second original region choice.

But why did AWS just move my system to a new region. That kinda bothers me.

Not sure if'll matter to anyone, but this was unusual enough that I wanted to let my fellow folks here know about it.
 
Both are pretty good. Used them both. Gcp is probably simpler to use (or at least it was when we got to use it) because it probably has lesser products to offer. You could go with either of them and have great and scalable application.
 
Well it depends pretty much on what you're trying the build. If your Backend is going to get bigger in the future you might want to take a look at AWS LAMBDA. They're serverless functions that allows a very friendly experience, because maintenance is painless. And if you connect your LAMBDAs to your repository through a pipeline then you would have CI/CD completed ;)

If it's just a simple website and you are looking to heavy use CDNs, then GCP is your choice along with Firebase libraries.

I wouldn't care that much which one performs better than the other, if you want high performance you should be more worried with how your code is built.

Regarding latency on regions, distance won't affect latency that much, unless your regions is in the other side of the world. I recommend using new regions available.
 
Advertise on BHW
You must log in or register to reply here.
Sign up now!

Main Menu

Home Main Forum List Black Hat SEO White Hat SEO BHW Newbie Guide Blogging Black Hat Tools Social Media Downloads

Marketplace

Account Selling Content / Copywriting EBooks / Methods Hosting Misc Proxies For Sale SEO - Link building SEO - Packages Social Media Social Media - Panels Web Design More

Making Money

Affiliate Programs Hire a Freelancer IM Journeys Making Money Pay Per Click Site Flipping Want to Buy

BlackHat World

Dispute Resolution Forum Suggestions & Feedback Introductions Lounge
Back
Top