[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[ih] vm vs. memory

On 10/24/2017 5:12 AM, Paul Vixie wrote:
> vm is an example of something that started as a workaround but 
> introduced us to a whole different way of thinking about memory. we now 
> have systems in production that always have physical RAM enough for 
> their work load, and who have no backing store for RAM (paging or 
> swapping) but which still depend on virtual memory for other reasons:
Agreed. The same could be said of NAT (created to save address space and
- IMHO - as a way for providers to enforce a pricing model that
differentiates service providers from clients).

Here's a summary of my interpretation of a useful view of NATs, i.e.,
that an "acid test" is that a NAT looks like a single host to the
external net and IMO ought to look like a router to the internal side:
J. Touch, ?Middlebox Models Compatible with the Internet
<http://www.isi.edu/touch/pubs/isi-tr-2016-711.pdf>,? USC/ISI Tech.
Report ISI-TR-711, Oct. 2016.

That's particularly useful when that's what you want (or need) to do,
e.g., to load balance the work of a single host on a bunch of internal
machines. It's painful when that's not what you want (or need) to do,
e.g., use a service someone incorrectly labels as "Internet access" but
is really "access to some content on the Internet".

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://elists.isoc.org/pipermail/internet-history/attachments/20171025/741d7c94/attachment.html>