Main fallacies when comparing your own servers with the cloud

Some use the clouds because of flexibility, other want to get away from routine, the third needs centralization, the fourth one needs security. However, the clouds are not only convenient, but also beneficial for a number of users. How competently to compare servers and clouds and not to make fallacies?

Consider the load limit for servers and storage systems

Practice shows that servers of the x86 architecture should not be constantly loaded by more than 70-80% on the processor and 80-90% on the memory, otherwise there may be performance drawdowns, especially noticeable during the start of maintenance tasks (for example, during backup). At the same time, short-term bursts of equipment load, as a rule, are normal, but with long-term operation in this mode, degradation of the performance of IT systems placed on this iron is very likely.

Consider the costs of key business systems`s sizing

Accurate planning of resources for key accounting information systems can be a difficult task, because each implementation is individual and often involves a large amount of refinements for specific business processes. In this case, when sizing, it remains to rely on the recommendations of system developers, who will certainly lay down possible risks - they do not want to justify themselves to the customer for the slow operation of the system after implementation. Having received the manufacturer's recommendations, their own IT specialists will also lay down the risks, because they do not want to listen to reproaches from users. As a result, it turns out that the manufacturer was reinsured when providing recommendations of times in half, and the IT department was reinsured when ordering equipment once and a half, and as a result, the purchased expensive iron is idle by more than half, and you can not take it back to the store.

It is necessary to take into account the cost of infrastructure maintaining

Any hardware, whether it's network hardware, servers or storage, needs to be serviced. The same can be said with regard to virtualization, monitoring, security, backup and other services ensuring stable and safe operation of the company's key business systems. The personnel, capable to support all this in working condition and repair in the event of an accident, is quite expensive. Often - unreasonably expensive for not the largest business.

At the same time, the total complexity of the stack of infrastructural technologies used is getting higher every year, and with it the requirements for the qualification of service specialists are constantly growing.