[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [cobalt-users] New to Internet servers
- Subject: Re: [cobalt-users] New to Internet servers
- From: Kris Dahl <krislists@xxxxxxxxxxxxx>
- Date: Wed Mar 29 11:12:22 2000
> For economic reasons I'm considering a 32mb/8.4ZGb Raq3. Is this option
> very limited? I plan to expand my client base to about 75-100 medium sites
> with 25-50 email users each. please comment.
Although its hard to evaluate bandwidth and server demands based upon the
'the number of sites', I am willing to bet you're not going to be happy with
the performance of that machine. *Especially* if you are doing light-medium
PERL and PHP scripting. The hard drive space doesn't matter all that much,
but I would recommend getting the 256M of RAM version.
> What are the advantages/disadvantages from IP based over Name based Sites?
> is there a best choice? does it make a difference?
Advantages of name based is: cheap(don't have to pay for IPs),
convenient(probably won't have to deal with ARIN), easy (don't have to
manage a large # of IPs.
Disadvantages: Some clients don't support it. Can suck if you have multple
hostnames pointing to the same name-based virtual host.
> Who do you recomend for co-locating?
Digital nation seems to be fairly popular. I have no first hand experience
with them.
> I would eventually like to set up a free web based email system for a
> medium sized city in mexico. Maybe about 1500 accounts to begin with. Would
> the Raq3 be able to handle it? what do you recommend?
The Raq could probably handle it, but I'd ask weather you'd be willing to
administer it. There is quite a bit that goes into the administration of a
site like that, including custom admin interfaces which usually nullify the
need for the Cobalt GUI. Typically companies that offer a service such as
this, have their own developers and hence are able to implement their own
system on more cost-effective servers.
I would be really hesitant to offer free or cheap anything hosting--you may
regret it at a later date (usually when its too late to do anything about
it)
I would, instead try to build a value added service that would make clients
WANT to pay for your services.
-k