The cloud storage business has turned into a cut-throat ruthless industry these days. So many cloud storage providers offer a huge amount of space in order to compete with their rivals. But it has been claimed that many cloud storage users waste nearly half of the cloud storage that they are paying for.
Traditional data centers only use 20%-30% of its capacity on average, so the cloud storage wastage statistic of 45% is not that far off. Cloud typically operates on a pay-per-use operating model, so users are left with the impression they are saving money compared to running a data center. But there are problems which contradict this logic.
Consider the following scenarios and you will see where the contradiction lies.
- Virtual machines are set up and running constantly, even though nothing is on them yet.
- Virtual machines are set up for temporary tasks but IT staff then forget to shut them down afterward. The worst case scenario then is you don’t know later which virtual machines can be shut down and which ones need to stay running.
- Virtual machines not being used at full capacity. IT staff over-estimate how much space is needed when in fact, a smaller amount of space would have been more than sufficient for the task at hand.
Using cloud services can also involve a huge amount of extra work. Whether it is custom scripts, or rewriting sections of an application’s code so it works with the cloud service, getting your company on board with the API’s of the cloud company can be a big ongoing task in itself. It also ensures that the costs of maintaining a cloud presence remain high.
What complicates the issue even more is that most organizations will use multiple cloud setups across different companies. This type of fragmentation is only going to make the situation less flexible and worse overall.
So what is the solution? How can you get the best bang for your buck from a cloud service? Quite simply by using “intelligent tools”.
One example is Cisco Cloudcenter, which offers various time-saving automated features, such as shutting down workloads that reach a certain age. Virtual machines can be scaled down when they are no longer needed. Different clouds can be compared to see where the best price savings are available.
Most important of all, virtual machines are only set up when an application needs to run. That virtual machine can then be shut down when the task is finished.
Cloud can be made to be cost-effective, but we first have to abandon the mindset that cloud can be used like a usual data center.