User comments on ISPs
  >> PlusNet plc


Register (or login) on our website and you will not see this ad.


These posts have been archived and can no longer be replied to or modified.
Pages in this thread: 1 | [2] | (show all)   Print Thread
Standard User rsturbolad
(regular) Sat 21-Mar-09 12:30:51
Print Post

Re: Packet shaping not as expected.


[re: deleted] [link to this post]
 
In reply to a post by 5km:
  • Web pages always allowed full line speed when web would work fine with 2Mbps.


Yes I've often wondered this; if you find a decent download site that isn't on the 'restricted list' or some other high content site, you could happily pull from it at line speed (for me the full 8meg) even at peak times - and it wouldn't take many doing that to cause problems.
Standard User deleted
(deleted) Sat 21-Mar-09 13:25:11
Print Post

Re: Packet shaping not as expected.


[re: Chrysalis] [link to this post]
 
Well one could argue that its as expected when there is a published specification and the throttling applied matches the spec. On the whole PlusNet seem to manage that quite well baring occasional mis configurations. Eg. one gateway un-throttled usenet a bit late last week on certain package(s). We can remember when it was a mess in their early days but they've worked out a fairly good way to configure the ellacoyas and make it understandable.

Its not trivial to understand though as the mechanisms that are simpler to explain are a) throttling your entire connection for all protocols (usually all the time), b) something like the original ALT which was effective and transparent and protected latency, c) letting congestion do its own thing. And maybe something like TalkTalk where p2p is throttled (but loads of people do not realise) but they never say by how much or when.

Enta's IPSC rollout has been far odder with certain connections showing signs of clear active throttling above the ALT, the ALT not behaving how it used to and its reporting being completely opaque when it used to be very transparent. And ofcourse a major shortage of capacity versus the packages sold and no resilience.

Now the OP may be right that some aspects of the PlusNet throttling specification are not how a techi might configure things BUT and I think this has always been the key point - to Joe Bloggs the internet means web browsing (and he may not even know what a browser is by name), if you can say that your browsing is unaffected and be able to prove it to the ASA then you are on a marketing winner. This is also useful for homes/businesses with multiple PCs (2Mb*4 = 8Mb) and so on. The main reason they throttle download sites (in my opinion) is simply that a) its easy b) they are by definition not real-time critical and c) when new releases appear its easy for them to flood the network. And with places like rapidshare its a), b) and a new c) which is that a user of those sites may be pulling a lot of Mb continuously.
Standard User deleted
(deleted) Wed 25-Mar-09 12:47:13
Print Post

Re: Packet shaping not as expected.


[re: deleted] [link to this post]
 
In reply to a post by 5km:
I've noticed that not all things are packet shaped as expected and how the current packet shaping doesn't make sense.

For example
  • Web pages always allowed full line speed when web would work fine with 2Mbps.


It is a misconception that web pages are not throttlled, often enforced by Plusnet reps, see my explanation here;

Link

I suppose it is easier to state the web is not throttled rather than explain the complex shaping.


Register (or login) on our website and you will not see this ad.

Pages in this thread: 1 | [2] | (show all)   Print Thread

Jump to