[Commotion-dev] Critical services

Tim Yardley yardley at gmail.com
Tue Feb 1 16:39:52 UTC 2011


Agreed if we get x86 hardware or equiv, but we might be limited to flash on COTS routers.

/tmy

On Feb 1, 2011, at 10:21 AM, Peter Folk <peter at volo.net> wrote:

> 
>> All of these sound like a good start.  The key is that localized
>> "caching" is limited by the hardware that we have (storage, etc).
>> With a flaky uplink, you want to be able to take an update and get it
>> where it needs to go whenever you have connectivity to do so.
> My guess is that the uplink from the gateway will be one of the most
> stable components of this whole system; just very slow.  That said the
> proxy I described is designed to help with flaky uplinks.
> 
> RE storage caching, it seems to me that even with relatively old
> hardware you can cache a lot: 10M would hold 10k compressed smtp
> messages or a similar number of RSS feeds.  I'm not talking about
> caching video, or even images, for the critical services.  That might be
> a reason to have the critical services proxy run on some gateways and
> the general http proxy run on others (so that it can use the max
> resources for caching).
> 
> Peter
> 
> 



More information about the Commotion-dev mailing list