There are enough variables involved in testing to make the result inaccurate. A couple of years ago I was testing a Hewlett-Packard Telco RMON using their NetMetics package. Since the monitor was attached directly to the telco circuit, the results were very accurate. The probe allowed you to selectively monitor performance on specific frame relay or multipoint connections.
It amazing how much the performance of the difference circuits varied. For example, circuits that went directly from the Frame-Relay POP to the site could burst to full T1 speed dispite the CIR whereas other circuits that went through multiple COs or across other transmission media (like microwave) were limited to the exact CIR. We also shared a T1 Internet connection with 2 other companies. The ISP "claimed" the three companies never used more than 7% of the bandwidth available dispite some frequent slowdowns.
Interesting enough, they were right! But the ISP bought bandwidth from another provider, who brought bandwidth from someone else who was connected to a major backbone provide (who at the time was having some problems of their own). The piering arrangement was what killed us. It was 10 hops just to get to an network access point like Mae West! Companies that used the same backbone provider were reasonably fast but if you wanted to get someplace else you were dead meat. Configuration, piering, congestion, and lots of other things effect performance. The only way IMO to get truely accurate results is to gather statics directly from the line.
-- Bill Stackpole, CISSP
| David Shimamoto <[EMAIL PROTECTED]>
Sent by: [EMAIL PROTECTED] 02/12/01 09:46 AM
|
To: Barry Hudson <[EMAIL PROTECTED]>, [EMAIL PROTECTED] cc: Subject: Re: Firewalls-Digest V8 #1479 |
I don't believe it is that accurate after running tests from desktops that
have 100 Mbps access out to core routers connected at DS-3 and higher speeds.
Some tests displayed less then T-2 speeds.
At 12:08 PM 2/12/01 -0500, Barry Hudson wrote:
>how accurate is this test?? I'm on a T1 and every time it says I'm off
>the chart above a T2.
>
>At 08:27 PM 2/9/01 +0000, you wrote:
>>Date: Fri, 09 Feb 2001 09:32:58 -0800
>>From: "Network Operations" <[EMAIL PROTECTED]>
>>Subject: Getting ripped off by your ISP?
>>
>>If you dont think you are getting the throughput promised you by your
>>ISP, here is a quick way to keep them honest! Check out:
>>
>>http://msn.zdnet.com/partners/msn/bandwidth/speedtest500.htm
>>
>>
>>(We have no interest whatsoever in the above site either personally,
>>financially or politically. I just remember several recent posts
>>requesting bandwidth testing)
>
>Barry S. Hudson
>Network Admin, CCNA
>Fredericksburg Savings Bank
>www.fsbnk.com
>Business Email - [EMAIL PROTECTED]
>All Other Email - [EMAIL PROTECTED]
>
>This email is intended for the addressee only. The material may be
>privileged and confidential information. If you have received this email
>in error, please notify me immediately by email and delete the original.
>Thank you.
>
>-
>[To unsubscribe, send mail to [EMAIL PROTECTED] with
>"unsubscribe firewalls" in the body of the message.]
-
[To unsubscribe, send mail to [EMAIL PROTECTED] with
"unsubscribe firewalls" in the body of the message.]
