Re: [RAM] The mapping problem: rendezvous points?

David Conrad wrote:
> Eliot,
>
> On May 9, 2007, at 9:13 AM, Eliot Lear wrote:
>> I feel like we've had this argument before.
>
> Yep.  You and others keep asserting that a pull-based mapping 
> distribution model is a "non-starter" because it "won't work".  I've 
> been trying to get from folks _concrete_ examples of applications that 
> "won't work" in the face of increased latencies on the order of 10s or 
> 100s of milliseconds after a cache miss.
I have a laundry list of issues with pull models, but they mostly 
evaporate if the host is involved.  I *like* DNS when the host is 
involved, for instance.  But when the device making the query *isn't* 
the host, things get messy for the application (discussed below).  Then 
there are security and continuity considerations.  Those are really fun, 
and they don't even involve the application.  Consider what happens when 
you have multiple middle devices that do this lookup and a major traffic 
path shifts.  How many queries will a middle device generate?  Can a 
reflection attack cause those queries to be generated?  These are 
concerns that have to be addressed.

> In response I get vague handwaving about voice or video (I know from 
> personal experience that UDP-based NFS works fine in the face of an 
> initial packet delay of 10s of milliseconds) or descriptions about 
> classes of applications that would appear to be degenerate cases and 
> asymmetric routing that would likely have unpredictable performance 
> characteristics with pull or push.
You talk delay.  I talk loss.  Routers don't just keep packets lying 
around if they can't deliver them.

That could change, but again it's a tradeoff.  I haven't tested v3 UDP, 
but earlier versions of UDP NFS suffer from a 90% throughput loss in the 
face of as much as 1% packet loss.  This is a well known fact due to its 
backoff algorithm (I used to work for an NFS server vendor in an NFS 
environment with devices that dropped packets all the time - painful).  
I'll leave it as an exercise as to what happens with voice and video 
quality when packets don't get delivered.

My message to you: we need to understand the tradeoffs of pull systems 
and not pretend they will come for free.

Eliot

_______________________________________________
RAM mailing list
RAM@iab.org
https://www1.ietf.org/mailman/listinfo/ram