headers
M. David Peterson | 1 Aug 2009 19:31

Re: Nginx is 119x faster than Cherokee 0.99.20 !

I'm just going to jump in here quickly and flat out call this the
lamest, most ignorant, and most ridiculous attempt at trying to make
it seem like Cherokee is not the web server it claims to be.

Firstly, do you have a fucking clue how to even run a benchmark.
Static files??? STATIC FILES!!!  What is this, 1996?

Secondly, as others have pointed out, do you honestly understand the
fact that running a benchmark on localhost will result in nothing but
completely useless data? You're not even touching the network card as
the requests are being sent and received via a loopback device, and
due to the heavy caching of files on the file system by the kernel,
you're likely not even hitting the hard disk after the first request.

Honestly, do you even know what goes into the reading, processing,
rendering, and transferring of a request before, during, and after a
request/response process has completed?  What you're doing, in
essence, is removing ALL of the bottlenecks that web servers have to
deal with during each transaction such as network latency, bandwidth
restrictions, disk I/O latency, compression/decompression, SSL
termination, reverse proxies, reading/writing/rewriting headers that
go beyond the typical headers found when requesting HTML, CSS,
Javscript, and/or other static files such as images, directory and
file handlers, security, etc. etc. etc. (and that list of etc.'s goes
on for MILES!)

The bottom line: Go read a fucking book about how hardware systems,
operating systems, network systems, and each and every one of the
application stacks that sit on top of them actually work. Then go and
write a benchmark based on a real world use cases that go beyond
(Continue reading)

Permalink | Reply | 12 comments |
headers
M. David Peterson | 1 Aug 2009 19:59

Re: Nginx is 119x faster than Cherokee 0.99.20 !

On Sat, Aug 1, 2009 at 11:31 AM, M. David Peterson<m.david <at> mdptws.com> wrote:

> I don't mean to slam you, man, but you should really consider building
> your street cred ^Up before attempting to slam the hard, life long
> work and dedication of others with FAR MORE street cred that you may
> ever gain (especially after crap like this) vDown. I won't have it. I
> can assure you of that.

BTW... I really hope you do take a step back and learn from all of
this as to the proper way to go about building up legitimate, well
deserved street cred in a world filled with unbelievably talented
folks from all sorts of open and proprietary software walks of life. I
know quite a few of the Apache developers, and have nothing but
MASSIVE respect for them. The same is true about the nginx developers,
the lighttpd developers, the IIS developers, and a list of million
people and 1000's of project long who deserve the upmost respect for
what they both have and continue to accomplish on a day to day basis
in their respective field.

You want an iconic figure to look up to and strive towards being even
half the developer he is:

http://www.google.com/search?q="Alvaro+Lopez+Ortega" < 26,400 entries,
each of which is related to something Alvaro has contributed in one
form or another to the advancement of computer science.  This man has
earned his respect in this industry through the sweat of his brow.

My guess is that you both can and will find your place and earn your
respect.  But the above approach is not the way.
(Continue reading)

Permalink | Reply | 0 comments |
headers
M. David Peterson | 1 Aug 2009 20:03

Re: Nginx is 119x faster than Cherokee 0.99.20 !

On Sat, Aug 1, 2009 at 11:31 AM, M. David Peterson<m.david <at> mdptws.com> wrote:

> ... which tells us what? That if we all would stop being so wasteful
> with how much markup and content we use then Cherokee would regain
> it's "rightful" spot as the /FASTEST/ completely useless static file
> server on the planet?

Uh, that should have read "Apache would regain it's rightful spot..." ;-)

--

-- 
/M:D

M. David Peterson
Co-Founder & Chief Architect, 3rd&Urban, LLC
Email: m.david <at> 3rdandUrban.com | m.david <at> amp.fm
Mobile: (206) 999-0588
http://3rdandUrban.com | http://amp.fm |
http://www.oreillynet.com/pub/au/2354 |
http://broadcast.oreilly.com/m-david-peterson/
Permalink | Reply | 4 comments |
headers
James Pearson | 1 Aug 2009 20:44
Picon
Gravatar

Re: Nginx is 119x faster than Cherokee 0.99.20 !

On Sat, Aug 1, 2009 at 11:31 AM, M. David Peterson<m.david <at> mdptws.com> wrote:

[snip rant]

First off, please realize that I am not disagreeing with you, but merely bringing up a few points.

Posting angry rantful messages usually does more harm than good in any internet discussion, a lesson that took me much too long to learn, and I still forget sometimes. :)  A simple "this is not realistic, for the following reasons" would have sufficed.  Giving an example of a good benchmark is always helpful, too.

A benchmark does not gain or lose credibility merely due to the experience of the person performing it.

Many people use handles for their work online (I do), so searching for their name will not bring up relevant results.

I doubt that anyone who subscribes to the cherokee mailing list will switch webservers because of a benchmark such as have been floating around the list recently.  If you're a subscriber, you're likely using cherokee in production, and will judge for yourself.

--
James Pearson
--
The best way to predict the future is to invent it.
 - Alan Kay
_______________________________________________
Cherokee mailing list
Cherokee <at> lists.octality.com
http://lists.octality.com/listinfo/cherokee
Permalink | Reply | 3 comments |
headers
Hello | 1 Aug 2009 20:59
Picon
Gravatar

Re: Nginx is 119x faster than Cherokee 0.99.20 !

hahahaha pwned and rightfully so

On Sat, Aug 1, 2009 at 2:44 PM, James Pearson <xiong.chiamiov <at> gmail.com> wrote:
On Sat, Aug 1, 2009 at 11:31 AM, M. David Peterson<m.david <at> mdptws.com> wrote:

[snip rant]

First off, please realize that I am not disagreeing with you, but merely bringing up a few points.

Posting angry rantful messages usually does more harm than good in any internet discussion, a lesson that took me much too long to learn, and I still forget sometimes. :)  A simple "this is not realistic, for the following reasons" would have sufficed.  Giving an example of a good benchmark is always helpful, too.

A benchmark does not gain or lose credibility merely due to the experience of the person performing it.

Many people use handles for their work online (I do), so searching for their name will not bring up relevant results.

I doubt that anyone who subscribes to the cherokee mailing list will switch webservers because of a benchmark such as have been floating around the list recently.  If you're a subscriber, you're likely using cherokee in production, and will judge for yourself.

--
James Pearson
--
The best way to predict the future is to invent it.
 - Alan Kay

_______________________________________________
Cherokee mailing list
Cherokee <at> lists.octality.com
http://lists.octality.com/listinfo/cherokee




--
Phone: 240-529-6770
Aim: flysarescary
_______________________________________________
Cherokee mailing list
Cherokee <at> lists.octality.com
http://lists.octality.com/listinfo/cherokee
Permalink | Reply | 2 comments |
headers
kevin beckford | 1 Aug 2009 22:08
Picon

Re: Nginx is 119x faster than Cherokee 0.99.20 !

> I don't mean to slam you, man, but you should really consider building
> your street cred ^Up before attempting to slam the hard, life long
> work and dedication of others with FAR MORE street cred that you may
> ever gain (especially after crap like this) vDown I won't have it. I
> can assure you of that.

Negated.  What the hell is wrong with you?  The guy had what he thought was an issue, posted to the list, asked for help.  We are free to question everything ( apparently as long as you are not running the show ) .   Why so nasty?  "Street Cred"  Are you serious?  Why not just say "Heresy!"

You are not an inquisitor, and this is not church.  I don't have the "street cred" of many on this list either, but I do understand civil discourse.  No gods are involved here, just humans.  

Now I have a question.  Are you speaking for yourself?  Or are you an attack dog, loosed to clumsily intimidate?  I can see no other reason for such vitrol.  

_______________________________________________
Cherokee mailing list
Cherokee <at> lists.octality.com
http://lists.octality.com/listinfo/cherokee
Permalink | Reply | 4 comments |
headers
Stefan de Konink | 1 Aug 2009 22:25
Picon
Gravatar

When I came here we were all friends

Can we please be friends again?

...because I like group hugs and parties. Not flamewars on mailinglist.
Anyone that want to have a kiss on the cheek :*

I like Cherokee, I like Alvaro, I like everybody here :) binb4sh you too!

If you don't like someone there is also a reply -not to list- button. But
keep in mind if many of us never responded on this list, or the bugtracker
we would still have a lot of bugs and don't spot the competition. We would
be friends and friends usually work with eachother and don't see what
happens around them.

Stefan
Permalink | Reply | 3 comments |
headers
Oli Warner | 1 Aug 2009 22:27
Gravatar

Re: Nginx is 119x faster than Cherokee 0.99.20 !

Mmmmkay

On Sat, Aug 1, 2009 at 6:31 PM, M. David Peterson <m.david <at> mdptws.com> wrote:
Firstly, do you have a fucking clue how to even run a benchmark.
Static files??? STATIC FILES!!!  What is this, 1996?

Static-file-serving is a major use of small servers like nginx, lighttpd and cherokee. Lots of people use them as a reverse proxy. Static performance is important. Step off.

 
Secondly, as others have pointed out, do you honestly understand the
fact that running a benchmark on localhost will result in nothing but
completely useless data?

Not completely useless. The benchmarks are consistent, in that tests on nginx and cherokee are both performed on localhost. If anything the processing power used by the benchmarking app only reduces the differences between the apps. It's all relative.

 
The bottom line: Go read a fucking book about how hardware systems,
operating systems, network systems, and each and every one of the
application stacks that sit on top of them actually work.

Perhaps he might when you've read a book on how to properly interact within an community. This is a mailing list for a web-server. Not a scientology congregation.

 
Please stop. You're not helping. You're hurting. So please. Stop.

Indeed.
_______________________________________________
Cherokee mailing list
Cherokee <at> lists.octality.com
http://lists.octality.com/listinfo/cherokee
Permalink | Reply | 0 comments |
headers
Oli Warner | 1 Aug 2009 22:31
Gravatar

Re: When I came here we were all friends

Indeed. Drama is a big waste of time.

*blushes* That said, I don't abide rude people. I just had a go in the thread of doom.

*hugs Stefan* Sorry!

On Sat, Aug 1, 2009 at 9:25 PM, Stefan de Konink <stefan <at> konink.de> wrote:
Can we please be friends again?

...because I like group hugs and parties. Not flamewars on mailinglist.
Anyone that want to have a kiss on the cheek :*

I like Cherokee, I like Alvaro, I like everybody here :) binb4sh you too!


If you don't like someone there is also a reply -not to list- button. But
keep in mind if many of us never responded on this list, or the bugtracker
we would still have a lot of bugs and don't spot the competition. We would
be friends and friends usually work with eachother and don't see what
happens around them.


Stefan


_______________________________________________
Cherokee mailing list
Cherokee <at> lists.octality.com
http://lists.octality.com/listinfo/cherokee

_______________________________________________
Cherokee mailing list
Cherokee <at> lists.octality.com
http://lists.octality.com/listinfo/cherokee
Permalink | Reply | 2 comments |
headers
Alvaro Lopez Ortega | 1 Aug 2009 22:43
Favicon
Gravatar

Re: Nginx is 119x faster than Cherokee 0.99.20 !

<SIGH..> I do know I shouldn't be replying to this thread. All these  
messages are obviously harmfull for the community. Besides, I'm well  
aware that this message will do nothing but keep this off-topic  
discussion alive..

On 01-ago-09, at 20:59, Hello wrote:

> hahahaha pwned and rightfully so

I don't think so. David raised very valid points, actually. In fact he  
basically wrote what I personally discussed with many other people  
after reading the original message.

By the way, if you are willing to accept my advice, perhaps you might  
want to start letting us know who you are. As David pointed, that  
allow the rest of us to know something about you; and that's  
important. Personally, I find interesting to know if the person  
usually works with other OSS projects (so he knows the basics), and  
what the rest of the say of him and his work [1].

Bare in mind, meritocracy is the base upon which the open source  
software is build upon. If the community knew how you are and what  
your previous work is, it'd be make interaction much easier for we  
all. For instance, in this case we have David, Eric and Hello. The  
first time I read David on the mailing list, I realized I knew him  
from somewhere else (a conference? another project? articles maybe?)  
so I searched his name on google. The result? Geez! I could have read  
the articles he writes at O'Reilly, it could also read about his work  
at Creative Commons, on XML, LLUP, Mono, or any other of the project  
he collaborates with. What did I think? "This is definitely a talented  
and motivated person. What an awesome new signing for the project!"

'Hello', do not take me wrong (no offense, seriously), but you can  
understand that an anonym users replying something like "hahahaha  
pwned" is far far away from that well deserved and trustful position.  
It's plain and simple meritocracy.

1.- http://www.youtube.com/watch?v=RrpajcAgR1E

> On Sat, Aug 1, 2009 at 2:44 PM, James Pearson <xiong.chiamiov <at> gmail.com 
> > wrote:
> On Sat, Aug 1, 2009 at 11:31 AM, M. David  
> Peterson<m.david <at> mdptws.com> wrote:
>
> [snip rant]
>
> First off, please realize that I am not disagreeing with you, but  
> merely bringing up a few points.
>
> Posting angry rantful messages usually does more harm than good in  
> any internet discussion, a lesson that took me much too long to  
> learn, and I still forget sometimes. :)  A simple "this is not  
> realistic, for the following reasons" would have sufficed.  Giving  
> an example of a good benchmark is always helpful, too.
>
> A benchmark does not gain or lose credibility merely due to the  
> experience of the person performing it.

Indeed. However, odds are an unexperienced person make mistakes, as  
we've seen in this case. Actually, experience is crucial for two kind  
of professionals: surgeons and engineers who perform benchmarks.. :-)

--
Octality
http://www.octality.com/
Permalink | Reply | 1 comment |

Gmane