Tech geek. Life geek.

Tag: performance (Page 1 of 3)

Update Samsung 970 EVO Plus NVMe M.2 SSD firemware

A few days ago, my desktop promoted that one firmware needs to be updated. It is a Samsung 970 EVO Plus NVMe M.2 SSD.  Almost nine months ago, I did a speed test on this SSD.

Now, after the update, I did the test again and see the difference.

First, the firmware update notice is on the Samsung Magician software.

Windows 10 notification

Update Samsung Firmware

So now the Firmware of My Samsung 970 EVO Plus NVMe 250GB SSD is updated from 1B2QEXM7 to 2B2QEXM7.

Drive Details of 970 EVO Plus

The first test is run on the old AS SSD Benchmark 1.8.5636.37293. All others are the same as before.

Continue reading

CDN compare: Azure vs Cloudfront vs BelugaCDN

It is a test of the speed from CDN Edge servers to the multi-locations. It is very important the signal for the real experience of the users.

Let me see how I implement the tests.

  • I created two endpoints on the Azure CDN, on Microsoft CDN and Akamai CDN.
  • I created a CloudFront distribution on Amazon CloudFront.
  • I created a CDN property on BelugaCDN.

All these four CDN settings pointed to a custom origin of one of my sites, which hosted on DigitalOcean.

The file I used for testing is a JPG file, an image file.

The testing tool is provided by KeyCDN performance tool.

I run at least two tests to make sure most of the testing server can get the JPG file from the cache of the Edge server(CDN), not back to the origin server.

 

Azure CDN: Standard Microsoft Tier

Standard Microsoft Tier CDN

 

The results are base on the 14 tests from different locations globally. I care about the TTFB much more than others.

What is TTFB

TTFB, which stands for time to first byte, is the amount of time it takes from when a client makes an HTTP request to it receiving its first byte of data from the web server. TTFB is an important aspect of website optimization since the faster the TTFB, the faster the requested resource can start being delivered to the browser.

The time to first byte is made up of three separate components.

  • The time needed to send the HTTP request
  • The time needed for the server to process the request
  • The time needed for the server to send back the first byte of the response to the client

What is a good TTFB?

The time to first byte can vary greatly depending on what sort of content you are serving (static vs dynamic), your server’s configuration, etc. Therefore determining what is a good TTFB number is difficult to answer and is variable based on your situation. However, on average anything with a TTFB under 100 ms is fantastic. Anything between 200-500 ms is standard, between 500 ms – 1 s is less than ideal and anything greater than 1 s should likely be investigated further.

As mentioned, it should be noted that these times are average benchmarks, however do not apply to all websites. Depending on the type of content and complexity of the application it may be completely unavoidable that the TTFB is greater than 1 second

Continue reading

AS SSD Benchmark Test on Crucial MX300 275GB

It is a quick test based on my Windows 10 Pro. It is just a new installation on my new desktop PC.

mx300-test

The testing tool is AS SSD Benchmark 1.9.5986.35387. It is latest one when I download it from the official site.

The result is not as good as I thought. I will check it later and see if it is any way can improve the performance.

Upgrade Curcial BX100 SSD Firmware

It is almost two years old of my Crucial BX100 256GB SSD. Now, I upgraded its firmware from MU01 to MU02. Then I did the benchmark test. The result is very clear that new firmware improve the performance.

The original test I did in 2014 shown as below:

sata3 ssd benchmark

AS SSD Benchmark test Result on Crucial BX100 256GB SSD with MU02 Firmware

The new test I did ten minutes ago.

ssd mu02 new firmware benchmark test

All test results are better than before. Same PC, Same SATA driver. Just different firmware.

Continue reading

PHP bench test results of PHP 5.3, PHP 5.4, and PHP 5.5

It is a bench test. Testing on the testing machine I just installed in the previous post.

Including PHP 5.3.28, PHP 5.4.28, and PHP 5.5.12, installed with fastcgi running on Apache and mod_fcgid.

PHP 5.3 is 6.567 seconds.

--------------------------------------
|        PHP BENCHMARK SCRIPT        |
--------------------------------------
Start : 2014-05-24 16:27:00
Server : web3.example.com@192.168.0.122
PHP version : 5.3.28
Platform : Linux
--------------------------------------
test_math                 : 1.724 sec.
test_stringmanipulation   : 1.741 sec.
test_loops                : 1.852 sec.
test_ifelse               : 1.250 sec.
--------------------------------------
Total time:               : 6.567 sec.

PHP 5.4 is 5.417 seconds.

--------------------------------------
|        PHP BENCHMARK SCRIPT        |
--------------------------------------
Start : 2014-05-24 23:27:23
Server : web2.example.com@192.168.0.122
PHP version : 5.4.28
Platform : Linux
--------------------------------------
test_math                 : 1.556 sec.
test_stringmanipulation   : 1.664 sec.
test_loops                : 1.328 sec.
test_ifelse               : 0.869 sec.
--------------------------------------
Total time:               : 5.417 sec.

 

PHP 5.5 is 5.646 seconds.

--------------------------------------
|        PHP BENCHMARK SCRIPT        |
--------------------------------------
Start : 2014-05-24 23:27:15
Server : web1.example.com@192.168.0.122
PHP version : 5.5.12
Platform : Linux
--------------------------------------
test_math                 : 1.618 sec.
test_stringmanipulation   : 1.723 sec.
test_loops                : 1.363 sec.
test_ifelse               : 0.942 sec.
--------------------------------------
Total time:               : 5.646 sec.

 

I Repeated the test many times and in different order. The results is consistent.

PHP 5.4 is a little bit faster than PHP 5.5. Both PHP 5.5 and PHP 5.4 are almost 1 second faster than PHP 5.3.

Performance between PHP5.3.3 and PHP5.5.12

I installed PHP5.3.3 and PHP5.4.28 at the same time in one VirtualBox guest OS.

Basic information.

  • Host OS: Windows 7 Ultimate 64bit
  • VirtualBox 4.3.12
  • Guest OS: CentOS 6.5 64bit with 512MB base memory
  • PHP5.3.3 is installed as mod_php
  • PHP5.5.12 is installed as php-fpm with mod_fastcgi

 

Run PHP BENCHMARK SCRIPT to get the total time.

php-benchmark-5.3.3

The total time of PHP 5.3.3 is 6.746 seconds.

php-benchmark-5.5.12

The total time of PHP 5.5.12 is 5.628 seconds.

I  repeated the test many times. The average difference is 85%. On the other words, PHP 5.5 spend 1 second less than PHP 5.3.

 

 

7 Tips to Wi-Fi Performance Optimization

The seven must-know tips for your Wireless Wi-Fi network Performance Optimization.

1. Placements of Devices

Radio signal travels through air to its destination. The radio signal will be reflected, refracted or absorbed if there is something on the way.

So, Keep the wireless router away from blockage, especially metal and bricks.

2. Antennas Positioning

Antennas has direction.  So try to put the antenna in multiple directions to avoid the generation of dead zone.  The D-Link SmartBeam provides much better directions.

3. Handling Radio Interference

As I discussed before, choose the least used wireless channel when possible.

4. Maximizing Channel Bandwidth

Later generations of wireless protocol may utilize higher bandwidth to deliver the superior performance from the previous. So choose 40MHz instead of regular 20MHz, it provides double bandwidth for your connection if possible with the wifi adapter.

5. Encryption

For security reasons, data encryption is must have. There are two main different encryption methods, or Cipher Type. TKIP and AES. TKIP has poor performance. So, select the AES for your network.

6. Other Unique Options

Some unique options may available for certain router and wireless adapters. Most are from same brand. Such as TX Bursting, Optimize AMPDU, Optimize ack suppression or Jumbo Pack.  But do it cautiously,  it may break the connection if the device is not compatible.

7. Further Extend the Wireless Coverage

To further extend the network coverage, one can also consider put a Wi-Fi repeater. I setup my WIRELESS-N WZR-HP-G300NH2 as repeater with DD-WRT rom.

Continue reading

How to Disable ETags in Apache httpd.conf

What are ETags?

ETag is a standard used for determining whether the client (web browser) already has the latest version. If it has, it’s simply loaded from local cache. If not, the new version is requested.

An ETag is a hash generated based on either the inode number, the last modified time and / or the size (in bytes) of a file. The default behaviour in Apache webserver is to use all properties to generate the ETag.

 

What’s the problem with ETags?

Although etags do no harm in a single server set-up, it is harmful when you have multiple servers replication the served files. The problem is that the generated ETags most likely differ per server. So when the client re-requests an un-updated file on a different server than the last time the ETags do not much and a new version is requested, while this would not be necessary. This hurts performance (page load). You can find more information about this in the ‘Configure ETags‘ article on the Yahoo Developer Network.

How To Disable ETags in httpd.conf

To disable the use of ETags in the Apache webserver simply put the following FileETag directive in your httpd.conf:

FileETag none

This will disabled the use of ETags completely.

 

4GB OpenVZ VPS

Ordered a new VPS from ServerMania. I did a test two days ago through ServerBear.

The basic information about this VPS.

  • RAM 4 GB
  • Burstable 8GB
  • HDD 150 GB
  • CPU Model Intel(R) Xeon(R) CPU E3-1270 V2 @ 3.50GHz
  • CPU Cores 4
  • CPU Speed 3500 MHz
  • CPU Cache 8192 KB
  • 4 TB Bandwidth
  • 4 IP address

The testing results:

UnixBench

UnixBench (w/ all processors) 3817.1
UnixBench (w/ one processor) 1900.7

Continue reading

« Older posts

© 2020 David Yin's Blog

Theme by Anders NorenUp ↑