Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

What exactly is the factor that differentiate/classify/categorize/determine between a Hosted datacenter and Cloud datacenter?

At an abstract level, if datacenter is a physical place where it is used to house (of course, air conditioning, maintenance, etc.) different customer servers, I wonder what is that primarily differentiates between a hosted and cloud datacenter?

share|improve this question
add comment

3 Answers

To some degree, it's marketing, so you'll see blurring of the lines between products. But in general:

-Cloud: You don't get access to the data center. You may not even know what city the data center is. You access the product through an API (that may be fronted by a control panel, etc.), and you may not get direct OS access, depending on the product. A definition that's increasingly used is that Cloud products are OSSM: On-demand, self-service, scalable, measurable. (https://twitter.com/wescpy/status/10165323856609280)

-Dedicated hosting: You don't get access to the data center, and you may not know exactly where it is. You lease a server (or a VM on a server), they probably put the OS on it for you (sometimes you can provide the OS for them to load, and sometimes you get KVM access to bare hardware where you can load your own OS). Probably the key differentiator is how automated the process is for the provider: As it tends more towards automated OSSM with an API, the provider is more likely to start marketing themselves as a "cloud" provider.

-Colo: They give you a rack with power and a cable or two giving you connectivity. You bring the equipment, and put it in your rack space.

share|improve this answer
add comment

A Colo has infrastructure in place to regulate access to specific banks of servers from arbitrary paying customers, where in a cloud (or shared) hosting DC that infrastructure is entirely virtual. That's the abstract answer. The Colo will have systems in place to monitor any number of things that will end up on a bill; the Colo will may be able to charge individual customers based on power usage, where the cloud/shared DC may just track power at the rack-level.

share|improve this answer
    
@sysadmin1138: I don't understand this A Colo has infrastructure in place to regulate access to specific banks of servers from arbitrary paying customers. Why there is a need to regulate access to specific banks of servers? –  Gnanam Feb 3 '11 at 13:57
    
@Gnanam: Because you wouldn't allow one customer to access another customers' equipment. –  joeqwerty Feb 3 '11 at 13:59
    
Well, you wouldn't want Joe Blow messing around with your gear, would you? –  gravyface Feb 3 '11 at 14:00
2  
At a cloud data center, there is no on-site access: the hardware layer is completely abstracted from you, as all your server instances are virtualized. Cloud computing is closer in similarity to a Virtual Private Server (VPS) than a traditional colocation facility. –  gravyface Feb 3 '11 at 14:36
1  
@Gnanam: At a cloud provider you don't have equipment, you have instances (of virtual machines). At a colo, you may have your own physical servers/rack and the colo operators don't let one customer access another customers' physical servers/rack. –  joeqwerty Feb 3 '11 at 17:04
show 1 more comment

The diiference exists at the business layer. What services are provided, do customers have their own equipment, do customers have access to their equipment, what services are available, etc., etc.

At the end of the day though, both models use physical hosts systems, routers, firewalls, air conditioning, fire suppression systems, etc., etc.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.