uptime of the host

Even the biggest and most redundant networks can't guarentee 100% uptime. You're kind of pitching your network against every disaster that could happen, even if the whole thing was hot-sited.
 
Even the biggest and most redundant networks can't guarentee 100% uptime. You're kind of pitching your network against every disaster that could happen, even if the whole thing was hot-sited.

this is true. A prime example is the recent Hurricane Sandy. this took down some of the largest Datacentres on the east coast which resulted in some major company sites going down.
 
It seems the competitive standard for uptime nowadays is 99.9%, which seems like a lot but should be capable for many host providers. If you want more than that, the only real solution is Cloud Hosting, which utilizes a network of servers to get the job done, which are capable of distributing new tasks to capable servers in the case that one should fail.
 
Is it not the case that these people are offering 99.9% uptime because if it does go down they know that the avarage customer won't be monitoring the downtime and working out the percentages.
 
Is it not the case that these people are offering 99.9% uptime because if it does go down they know that the avarage customer won't be monitoring the downtime and working out the percentages.

No, offering 99.9% can be acheived and the way we look at it is that we are the ones that should be monitoooring our servers and not our clients for downtime. we will always compensate clients for downtime even if they dont notice as there sites were still down regardless.
 
Is it not the case that these people are offering 99.9% uptime because if it does go down they know that the avarage customer won't be monitoring the downtime and working out the percentages.

Not really. The main driving is that the marketing advantage of that claim compensates for the cost of the backing up the promise. Let me explain what I mean.....

Suppose I charge $3.00/mo for hosting with a 100% uptime gurantee. Sounds good, doesn't it? Now let's suppose my customers experience 90 minutes of downtime in one monthly period. That's 0.2% downtime; my promised 100% uptime was not achieved -- I delivered "only" 99.8% uptime.

I need to refund at least a prorated amount. That would be only 6 cents! In other words every monthly 0.1% of downtime would cost me 3 cents per month per customer (that claims the refund).

Thus, a host can exaggerate uptime guarantee while planning for refunds. Moreover, these contemplated refunds can be built into the price. This can give rise to a case where more refunds = more profits

See also my post on the SLA http://www.hostingdiscussion.com/cu...ime-guarantee-good-bad-idea-5.html#post159106
 
I think 99% + is already pretty good, because there must be times that it would be down...it is even possible that the servers are down for 1 sec. ... may be, just assuming.

You know! So yah...100% is definitely achievable, but difficult to.
 
What is an average server uptime of the greater part of the hosts?
Share your experience please.

A lot promise 100% uptime but we know that is impossible, they have to rely on giving SLA when there is an issue.

That's why we give 99.9% uptime, and a lot of hosts do too. Much more reasonable :)
 
Back
Top