How do you know if cloud computing makes sense for your organization? And if it makes sense for your organization what sort of services should you buy into? Even though cloud computing is everywhere, is being offered by all the major technology players, and all agree you need it, definitions vary.
Cloud Computing refers to the delivery of infrastructure components and services. Its name is derived from architecture diagrams where a cloud is used as a metaphor for the Internet. A cloud is used to abstract and represent the complex infrastructure environment.
The components that are housed in the cloud on the diagram are client devices, servers and data centers. Client devices include mobile devices, think and thick computers. While the data center in the cloud are defined as servers or collections of servers, for small organizations the cloud may represent a portion of a server. Being able to provide a portion of a server, or storage space, called ‘capacity on demand’ is one of the hallmarks of cloud computing.
The ‘as a service paradigm’ in cloud computing has vendors offering Software as a Service (SaaS), Infrastructure as a Service (IaaS), Database as a Service (DaaS), Platform as a Service (PaaS), and more. In addition to getting the right balance of cost to benefit, availability, scalability, capacity, and regulatory concerns will drive the decision process.
Additionally, determining your organization’s risk profile will be key to understanding if cloud computing and associated service is right for you. So what service makes sense for your organization? Here are the questions to ask.
1. Does the cost benefit justify the disruption?
Data center migrations are disruptive, costly, and complicated. Tightly coupled mission critical systems that make up the minority of the data center take the most time and expense to migrate. Though savings range from project to project and will vary by organization, the greatest savings will be found with labor, hardware, and software. The key is that the cost benefit outweighs the disruption the migration will cause.
2. Do your applications require specific hardware components or speed of delivery?
Some applications cannot be virtualized as they require specific underlying hardware components. While tempting, it is best to not migrate these to the cloud. Additionally, some applications require minimal delivery speeds. If latency is an issue, it is best to keep the entire system (i.e., all the system components - web, application, database, middleware servers, etc.) local. Simply look for other cloud opportunities within your environment (e.g., end of life servers hosting non-critical applications).
3. How much capacity will you use?
While you own or lease your servers the capital outlay is quite consistent from month to month. Typically, if more capacity is required, someone creates a purchase order and a while later new assets are available. Because cloud computing is an ‘on demand’ service, scalability is provided as needed. Which is a good thing, right? Maybe.
Some IT shops may find their costs spiking as utilization increases; if this was planned, it is not an issue, but best to do the math before signing up.
4. Are data regulation and security hurdles too high to overcome?
Not all data can or should reside online. Your IT security group or governing body may mandate that sensitive data remain local. Regulatory constraints aside, from a security perspective the organization has to be willing to risk placing their and their customer’s data in the hands of a third party vendor.
The organization has to be willing to give up some level of control because the cloud is like a black box. Banks and health care organizations may find this an unacceptable risk, regardless of the controls that a vendor may employ.
5. When shouldn’t I use cloud computing?
- If you have Health Insurance Portability and Accounting Act (HIPAA) data, you should not. The last thing you want is your data to co-mingle with someone else’s. Better not to take a chance.
- Sensitive data is best kept local. Sure you can use encryption, but with any good security paradigm you make it more difficult for hackers, bots and yourself.
- If your applications require specific hardware components, or your application requires complete access to the server, cloud computing may not be a good fit.
- Since data and applications may be in disparate locations, if latency is an issue, then the cloud is not a good fit.
- Tightly coupled, mission critical applications are not good candidates for cloud computing.
- Cloud inefficiency - Bad habits are hard to break
- Disaster recovery needs leading SMBs to cloud
- The curse of free cloud services: a cautionary tale
- How the cloud helps build agile companies and encourage experimentation
- Build your own open-source cloud with ownCloud 4
- How to calculate what to move to the Windows Azure cloud