The cloud is the latest fad and for a good reason too. It offers the advantages of scalability, resilience, redundancy and more. However, for all its advantages, there are many myths that have cropped up as well. The following are the three top myths associated with cloud computing:
Myth 1: The cloud is not secure.
The very mention of cloud computing resonates with security risks. There is some truth in this myth. Public cloud providers adopt a multi-tenant model to justify economies of scale and this can compromise security. Moreover, the client loses control of the data to a third party cloud provider. The Cloud Security Alliance lists many security risks such as abuse by malicious insiders, vulnerabilities resulting from shared technologies, risk of data loss or leakage, account/traffic hijacking, insecure application programming interfaces, etc. as prevalent with cloud computing.
However, the concerns of security are vastly exaggerated. Traditional on-premises have the network firewall as the security perimeter. In cloud, the hypervisor and/or the underlying cloud application become this security perimeter. A solid design combined with operational rigor which makes security a priority can offer top grade security for cloud accounts.
Proper internal controls forced upon by the need to transfer data to and from the third-party cloud network may mean that the cloud may offer better security than on-premises computing. The game changer is getting this right.
The cloud client may also ensure proper visibility to the cloud network through HP Cloud Assure or some other similar application.
Myth 2: Critical applications do not belong in the cloud
An oft-touted strategy to mitigate the security risk associated with the cloud is refraining from migrating critical applications to the cloud. While this may improve security, it very often defeats the very purpose of embracing cloud in the first place. After all, what significant business advantage can one achieve by relegating a few servers running test and development jobs to a cloud-based infrastructure, leaving the business applications out?
Of course there are several ways to have the best of both worlds. They are:
- Enlist a reputed and credible provider. The cloud essentially transfers the security risks from the client end to the cloud provider’s end. As long as the cloud provider does a good job with security and data encryption is in place, there is little cause for concern
- Adopt private clouds or hybrid cloud models which improve security while retaining the other benefits of the public cloud
- Ensure the accurate deployment, configuration and sizing of cloud services using applications such as HP Cloud Maps
Myth 3: The public cloud is inexpensive than on-premises computing
Cloud providers tout the cost savings that result from the “pay-as-you-go” model, especially the ability to set up computing infrastructure without considerable upfront infrastructure as a big advantage of migrating to the cloud. The starting price for the standard On-Demand Instances of Amazon EC2 Web service, for instance, costs about a dime an hour.
Cloud computing leverages economies of scale and some of its benefits are passed on to the client. Moreover, the cloud breaks down large upfront expenditures into ongoing operational costs to offer pricing with usage. This apparently means that the clients end up paying less than what they would if they have to invest in their own infrastructure.
There is, however, more to this than meets the eye. The cost effectiveness works only when the client requires the resources sparingly or intermittently. For resources required on a continuous basis, shared resources via private clouds may work out to be more cost-effective without compromising on any of the benefits of the cloud.
To put it simply, the cost effectiveness of the cloud is a part of the quintessential rent vs. buy conundrum. The guiding factors on whether it makes sense to adopt a pay-as-you-go model or invest in your own infrastructure are no different than the guiding factors behind a decision to hire a car or buy one.
At another plane, many people confuse price with value. Critical issues such as uptime availability, performance, security, compliance, support and service-level agreements need factoring in along with the raw price to determine the best value-for-money preposition.
There is no one-size-fits-all solution when it comes to the best cloud computing model. While the cloud is the latest fad and has its own uses, it is foolhardy to paint it as the “best” model and the panacea to all technology-related business problems.