A cloud servers Vietnam DC has incredibly high hypervisor control panels scaling abilities, but not many of them realize its real capacity.
Have you ever wonder as to how good the performance of cloud as I/O utilization, storage, networking and storage really is and the number of the virtual machines it is capable of hosting? Today, cloud data centers are bulked up, big, and in fact, really pumped up, but what’s their real strength?
In terms of technic, do they have the ability to perform or can they scale? The answers are not too straightforward.
Assessing the real capacity – a tricky task
Knowing a data center’s capacity is almost impossible. The technical specifications may seem to be easy, but they do not really measure the end-to-end performance of the whole stack.
In many organizations, the engineers themselves aren’t aware of the number of VMs that their cloud can host. The statistics of the real and virtual CPU and I/O utilization are usually monitored by the operations department and when the pre-set thresholds are reached the load is redistributed or more resources get added.
Actually, the engineers aren’t aware of the cloud capacity of the DC in terms of how software/ hardware stack of the cloud are tuned for maximum scalability, where performance hold-ups are and traditional capacity racks have for growth in future.
They’re mostly aware of only the capacity figures like number of CPUs, number of servers, network bandwidth, and storage capacity. They lack of the knowledge on whether the cloud can get provisioned by a single gear rack or dozens and aren’t confident of their space estimates.
Specs do not indicate everything
The primary aspect of such an issue is that it’s not easy to decipher this. The specification sheets do not indicate anything; nor do the VM hypervisor control panels. For big cloud infrastructure, its capacity depends on the best presumptions worked out in excel, further depending upon extrapolating a single computer’s workload. However, that is not the way in which the cloud scales, especially when you consider the multi-software, hardware and infrastructure levels in the stack.
You need to factor in the dynamic architectures provided by Software Defined Networks, and software-oriented firewalls and load-balancers described by Network Functions Virtualization. Nobody knows the outcomes, which means everyone’s guessing things following which they observe the operations in real-time to handle real-time capacity.
They hope that the admins will observe the slowdown in performance with adequate time space to respond before there are any outages. Even in such cases, they should provide a fast response. This does not help with capacity planning for the long-term, which is why it is necessary to have plenty of extra capacity prepared to enter the cloud era within short notice.