From: "Robert P. J. Day" on
Quoting Jochen Schultz <jschultz(a)sportimport.de>:

> OS?
>
> Robert P. J. Day schrieb:
>> probably not really a PHP question but i'll take a chance, anyway.
>> i want to examine the network throughput i can get when
>> continually uploading files from a PHP script via a POST request
>> using the HTTP_Request2 class.
>>
>> i have a client-side script that simply takes files, creates a
>> short POST request, and submits it to a server-side PHP script that
>> takes the uploaded file and saves it. no big deal.
>>
>> as a test, i created a random 5M file, then looped 100 times
>> submitting the same file, and timed it. while the system and user
>> time was only a few seconds total, real (clock on the wall) time
>> was 2.5 minutes. this suggests that the bottleneck is simply
>> network transfer speed.
>>
>> while i'm doing these uploads, is there a way to monitor network
>> throughput? if this is truly the bottleneck, the only real
>> solution will be to pay a premium for faster network access, i
>> suppose. but i'd just like to be able to produce some numbers or
>> evidence that that's the actual problem. thoughts?

whoops, sorry, CentOS 5.4.

rday

From: tedd on
At 6:52 AM -0400 5/5/10, Robert P. J. Day wrote:
> probably not really a PHP question but i'll take a chance, anyway.
>i want to examine the network throughput i can get when continually
>uploading files from a PHP script via a POST request using the
>HTTP_Request2 class.
>
> i have a client-side script that simply takes files, creates a
>short POST request, and submits it to a server-side PHP script that
>takes the uploaded file and saves it. no big deal.
>
> as a test, i created a random 5M file, then looped 100 times
>submitting the same file, and timed it. while the system and user
>time was only a few seconds total, real (clock on the wall) time was
>2.5 minutes. this suggests that the bottleneck is simply network
>transfer speed.
>
> while i'm doing these uploads, is there a way to monitor network
>throughput? if this is truly the bottleneck, the only real solution
>will be to pay a premium for faster network access, i suppose. but
>i'd just like to be able to produce some numbers or evidence that
>that's the actual problem. thoughts?
>
>rday

You might want to test your connection, such as:

http://www.speakeasy.net/speedtest/

Cheers,

tedd


--
-------
http://sperling.com http://ancientstones.com http://earthstones.com