how-to block ads
|
anonagain
@candlelight.ca
| reply to jfmezei
Re: Simple document to help explain the issues
With the greatest respect, I think you may be mistaken.
Having read the tariff (filed with the CRTC and pretty close to the bible in any dealings based on a tariff structure) carefully, it looks like the tariff actually describes AHSSPI as being a "burstable" service.
Quote: "1000 Mbps (Burstable up to 1000 Mbps) per service provider / 1000 Mbit/s," is the language in the tariff.
Since "burstable" generally translates to "oversubscribed" in network-provisioning-speak, is it not actually possible the Bell has been caught with their pants down and really does have a problem with congestion at the core network level with AHSPPI?
Please consider »www.bce.ca/en/aboutbce/regulator···20%20%20
Just a thought, as it seems like there is a lot of discussion about what should be rather than what is actually printed on the tariff that Bell is bound by.
I would be very interested to hear Rocky's point of view here as it certainly seems to me that Bell has all kinds of wiggle room to justify their congestion argument, frivolous though that argument may be.
George | |
Maynard G Krebs
@teksavvy.com |
It would all depend on what Rocky's contract stated. | |
anonagain
@candlelight.ca
| Just so.
And that would likely depend on when it was signed, and perhaps what the tariff language was at the time. I have no idea what would rule in court, the tariff or the contract.
I see that DS-3 and OC-3 are fixed rate, but all the Ethernet delivery options are burstable.
To my mind, that's a potential issue.
George | |
jfmezei
Premium
join:2007-01-03
 ·TekSavvy Solutions..
| reply to anonagain
Quote: "Quote: "1000 Mbps (Burstable up to 1000 Mbps) per service provider / 1000 Mbit/s," is the language in the tariff."
There are no descriptions of what an acceptable sustainable rate for a 1gbps link is to be.
The fee structure doesn't take into account actual usage, it is a fixed fee. Some services give you a 1gbps pipe, but only charge you for what you actually use (eg: measure the bursts). In such cases, your monthly bill can vary if your users change their patterns significantly.
So since the AHSSPI doesn't specify acceptable average utilisation rates, nor do the fees vary, then the term "burstable" doesn't have much of a meaning.
Also, cosnider a case where an ISP buys 4gpbs capacity, but because of the way AHSSPI is built, it requires 4 separate 1gps feeds. If you burst your second link a lot, but all others remains well below the 1gbps mark, how would it be billed if it were a variable charge ? | |
anonagain
@candlelight.ca
| said by jfmezei9 
There are no descriptions of what an acceptable sustainable rate for a 1gbps link is to be.
The fee structure doesn't take into account actual usage, it is a fixed fee. Some services give you a 1gbps pipe, but only charge you for what you actually use (eg: measure the bursts). In such cases, your monthly bill can vary if your users change their patterns significantly.
So since the AHSSPI doesn't specify acceptable average utilisation rates, nor do the fees vary, then the term "burstable" doesn't have much of a meaning.
Also, consider a case where an ISP buys 4gpbs capacity, but because of the way AHSSPI is built, it requires 4 separate 1gps feeds. If you burst your second link a lot, but all others remains well below the 1gbps mark, how would it be billed if it were a variable charge ?
[/BQUOTE : You hit on the issue. There is absolutely no description about what you get when you buy a 1 gig burstable link. None. Zero. Nada. I've been on both sides of the debate long enough (about 36 years as a customer or a provider give or take) to know that when there is vague language, whoever has the most, and most expensive lawyers tends to win. Contracts with Sprint, AT&T, Bells of all colours, MCI, Worldcom, Quest, NTT, BT etc. "Burstable" is one of the best get-out vague descriptions there has ever been. It clearly means you don't get a gig full time, but it doesn't guarantee what you do get. Good contracts specify a CIR/MIR/95% percentile, plus they specify a service level agreement for getting 'stuff' fixed. Further, they define what 'stuff' actually means. And they specify who is monitoring the 'stuff', how it is monitored, and what the dispute resolution mechanism is. In other words, the tariff sucks. I really hope this doesn't torpedo the whole claim with Bell. As Bell sucks anyway. I would love to hear the exact contract language vis a vis the tariff... George | |
|
