This is a poor benchmark because the query never changes. Query cache takes 
over after first request. 

Sent from my iPad

On Aug 16, 2011, at 4:28 PM, "Alvin Ramos" <alvin.ra...@reachsmart.com> wrote:

> I know my previous email was vague, it was sent via smartphone.  I’ve got a 
> simple PHP page pulling information from one of larger database tables:
> 
> 
> 
> PHP Code:
> 
> 
> 
> <html>
> 
> <head>
> 
> <basefront face="Arial">
> 
> </head>
> 
> 
> 
> <body>
> 
> 
> 
> <?php
> 
> 
> 
> // set server access variables
> 
> $host = "127.0.0.1";
> 
> $user = "web";
> 
> $pass = "password";
> 
> $db = "md ";
> 
> 
> 
> // open connections to database
> 
> $connect = mysql_connect($host, $user, $pass) or die ("Unable to connect!");
> 
> 
> 
> // select database to use
> 
> mysql_select_db($db) or die ("Unable to select database!");
> 
> 
> 
> // create SQL query string
> 
> $query = "SELECT * FROM members limit 1000";
> 
> 
> 
> //execute query and obtain result set
> 
> $result = mysql_query($query) or die ("Error in query: $query. " . 
> mysql_error());
> 
> 
> 
> // are there any rows in the result?
> 
> if (mysql_num_rows($result) > 0)
> 
> {
> 
>        // yes
> 
>        // iterate through result set
> 
>        // format query results as table
> 
>        echo "<table cellpadding=10 border=1>";
> 
>        while($row = mysql_fetch_assoc($result))
> 
>        {
> 
>                echo "<tr>";
> 
>                echo "<td>" . $row['member_id'] . "</td>";
> 
>                echo "<td>" . $row['fname'] . "</td>";
> 
>                echo "</tr>";
> 
>        }
> 
>        echo "</table>";
> 
> }
> 
> else
> 
> {
> 
>        // no
> 
>        // print status message
> 
>        echo "NO rows found!";
> 
> }
> 
> 
> 
> // close connection
> 
> mysql_close($connect);
> 
> 
> 
> ?>
> 
> 
> 
> </body>
> 
> </html>
> 
> 
> 
> I’ve got apache benchmark then running 5 concurrent connections 10,000 times. 
>  I changed the $host to the IP for the 5.5 server then to the 5.1 server and 
> here are one of my many results:
> 
> 
> 
> 5.1 results:
> 
> 
> 
> Server Software:        Apache/2.2.3
> 
> Server Hostname:        aramos.dev
> 
> Server Port:            80
> 
> 
> 
> Document Path:          /mysqlfetch51.php
> 
> Document Length:        35808 bytes
> 
> 
> 
> Concurrency Level:      5
> 
> Time taken for tests:   3263.909079 seconds
> 
> Complete requests:      10000
> 
> Failed requests:        0
> 
> Write errors:           0
> 
> Total transferred:      359640000 bytes
> 
> HTML transferred:       358080000 bytes
> 
> Requests per second:    3.06 [#/sec] (mean)
> 
> Time per request:       1631.955 [ms] (mean)
> 
> Time per request:       326.391 [ms] (mean, across all concurrent requests)
> 
> Transfer rate:          107.60 [Kbytes/sec] received
> 
> 
> 
> Connection Times (ms)
> 
>              min  mean[+/-sd] median   max
> 
> Connect:        0    1   2.3      1     155
> 
> Processing:   593 1629 699.7   1524   13580
> 
> Waiting:      574 1611 699.7   1506   13562
> 
> Total:        595 1630 699.7   1526   13580
> 
> 
> 
> Percentage of the requests served within a certain time (ms)
> 
>  50%   1526
> 
>  66%   1725
> 
>  75%   1856
> 
>  80%   1944
> 
>  90%   2215
> 
>  95%   2559
> 
>  98%   4339
> 
>  99%   4741
> 
> 100%  13580 (longest request)
> 
> 
> 
> 5.5 results:
> 
> 
> 
> erver Software:        Apache/2.2.3
> 
> Server Hostname:        aramos.dev
> 
> Server Port:            80
> 
> 
> 
> Document Path:          /mysqlfetch.php
> 
> Document Length:        35808 bytes
> 
> 
> 
> Concurrency Level:      5
> 
> Time taken for tests:   3400.300474 seconds
> 
> Complete requests:      10000
> 
> Failed requests:        0
> 
> Write errors:           0
> 
> Total transferred:      359640000 bytes
> 
> HTML transferred:       358080000 bytes
> 
> Requests per second:    2.94 [#/sec] (mean)
> 
> Time per request:       1700.150 [ms] (mean)
> 
> Time per request:       340.030 [ms] (mean, across all concurrent requests)
> 
> Transfer rate:          103.29 [Kbytes/sec] received
> 
> 
> 
> Connection Times (ms)
> 
>              min  mean[+/-sd] median   max
> 
> Connect:        0    1   2.7      1     168
> 
> Processing:   595 1697 724.8   1598   14505
> 
> Waiting:      577 1679 724.8   1580   14486
> 
> Total:        596 1698 724.8   1600   14506
> 
> 
> 
> Percentage of the requests served within a certain time (ms)
> 
>  50%   1600
> 
>  66%   1799
> 
>  75%   1939
> 
>  80%   2028
> 
>  90%   2314
> 
>  95%   2640
> 
>  98%   4387
> 
>  99%   4805
> 
> 100%  14506 (longest request)
> 
> 
> 
> 
> 
> I’ve ran tests even against our web sites and its slower than the 5.1 server. 
>  Any suggestions, anything I should change on the 5.5 server?  The hardware 
> and OS is identical from the 5.1 server.  Thanks!
> 
> 
> 
> Alvin Ramos
> 
> 
> 
> From: w...@pythian.com [mailto:w...@pythian.com] On Behalf Of Singer X.J. Wang
> Sent: Tuesday, August 16, 2011 4:08 PM
> To: Alvin Ramos
> Cc: Prabhat Kumar; Reindl Harald; mysql@lists.mysql.com
> Subject: Re: shall i jump from 5.1 to 5.5
> 
> 
> 
> Are you doing concurrent workloads?
> 
> 
> 
> On Tue, Aug 16, 2011 at 16:04, Alvin Ramos <alvin.ra...@reachsmart.com> wrote:
> 
> I've been running some bench marking between 5.1 and 5.5 myself and haven't 
> notice any huge performance improvements on 5.5. Even though white papers 
> claim it put performs 5.1. Any noticing the same or have some input in my 
> findings?
> 
> Regards,
> Alvin
> 
> 
> On Aug 16, 2011, at 3:55 PM, Prabhat Kumar <aim.prab...@gmail.com> wrote:
> 
>> correct. you have to understand the problem first.
>> but still its recommendable to always use latest stable version.
>> 
>> On Tue, Aug 16, 2011 at 10:02 AM, Reindl Harald 
>> <h.rei...@thelounge.net>wrote:
>> 
>>> 
>>> 
>>> Am 16.08.2011 17:59, schrieb Luis Daniel Lucio Quiroz:
>>>> as far as my readings, they claim that 5.5 is the best
>>>> 
>>>> my question is, shall i jump from 5.1 to 5.5.
>>>> 
>>>> right now i have a performance problem, would 5.5 help me in that?
>>>> 
>>>> Regards,
>>>> 
>>>> LD
>>>> 
>>> why do you believe without any information you will get
>>> a useful answer? "i have a performance problem" is simply
>>> NO information if you even do not tell which storage engine
>>> and wich sort of problem in which context
>>> 
>>> if you should update can nobody answer for you because we
>>> do not know if you have any crappy apps / scripts which
>>> would have troubles?
>>> 
>>> we have upgraded some hundret webspaces and two dbmail-servers
>>> in februray becaus we know our self written applications and
>>> having test-environments, if you can do this can nobody say
>>> 
>>> 
>> 
>> 
>> --
>> Best Regards,
>> 
>> Prabhat Kumar
>> MySQL DBA
>> 
>> My Blog: http://adminlinux.blogspot.com
>> My LinkedIn: http://www.linkedin.com/in/profileprabhat
> 
> 
> 
> --
> MySQL General Mailing List
> For list archives: http://lists.mysql.com/mysql
> To unsubscribe:    http://lists.mysql.com/mysql?unsub=w...@singerwang.com
> 
> 
> 
> --
> The best compliment you could give Pythian for our service is a referral.
> 

--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:    http://lists.mysql.com/mysql?unsub=arch...@jab.org

Reply via email to