Cloud Computing Isn't For Everyone
Companies need to make sure they understand the security risks. Cloud computing has evolved from a truly bizarre concept to a mainstream buzzword.
Nov 12, 2008
Cloud computing has evolved from a truly bizarre concept to a mainstream buzzword.
The idea of handing over the keys to corporate data sounded almost ludicrous when it was first proposed several years ago. Since then, big companies like IBM (nyse: IBM - news - people ), Salesforce.com (nyse: CRM - news - people ) and Amazon (nasdaq: AMZN - news - people ) have jumped on the bandwagon and all are pitching cloud computing solutions. With giants like that behind it, surely it can't be all bad, right?
The truth is, it all depends. There are big caveats associated with this model. Sometimes it works, sometimes it doesn't. For new companies, the decisions are easier. It's a question of renting versus buying. If the value is in the service rather than the infrastructure, then this model can make sense. Why buy servers and hire people to run back-end services when you can simply outsource the whole operation and pay as you go? The whole idea behind cloud computing is that you don't care where the data is going and where it's being processed.
Cloud computing means a company's IT staff doesn't have to gain expertise in a new application. And since IT staffs are already stretched pretty thin these days, companies are probably going to do a better job of running the application and updating it. You can set up rules for someone hosting an application, but you can't set rules for a cloud model.
For more established companies, pulling the trigger on this concept is a lot more complicated. While it may be cost effective to outsource certain applications, it isn't necessarily true that it makes sense to run them in a cloud.
This is easier to figure out in established companies than it is in start-ups, and the older the company and the more layers of IT that have been added over the years, the more difficult it is to ascertain what connects to what. It's also necessary to figure out what's valuable inside a company and where the access points are to that data.
The idea of handing over the keys to corporate data sounded almost ludicrous when it was first proposed several years ago. Since then, big companies like IBM (nyse: IBM - news - people ), Salesforce.com (nyse: CRM - news - people ) and Amazon (nasdaq: AMZN - news - people ) have jumped on the bandwagon and all are pitching cloud computing solutions. With giants like that behind it, surely it can't be all bad, right?
The truth is, it all depends. There are big caveats associated with this model. Sometimes it works, sometimes it doesn't. For new companies, the decisions are easier. It's a question of renting versus buying. If the value is in the service rather than the infrastructure, then this model can make sense. Why buy servers and hire people to run back-end services when you can simply outsource the whole operation and pay as you go? The whole idea behind cloud computing is that you don't care where the data is going and where it's being processed.
Cloud computing means a company's IT staff doesn't have to gain expertise in a new application. And since IT staffs are already stretched pretty thin these days, companies are probably going to do a better job of running the application and updating it. You can set up rules for someone hosting an application, but you can't set rules for a cloud model.
For more established companies, pulling the trigger on this concept is a lot more complicated. While it may be cost effective to outsource certain applications, it isn't necessarily true that it makes sense to run them in a cloud.
This is easier to figure out in established companies than it is in start-ups, and the older the company and the more layers of IT that have been added over the years, the more difficult it is to ascertain what connects to what. It's also necessary to figure out what's valuable inside a company and where the access points are to that data.