A better understanding of the cloud requires to know the definitions being used.
We'll give you an overview of the most frequently used terms with some useful notes about what you are about to buy.
It may reference to the virtual CPU (vCPU), to the virtual cores of the CPU, to the dedicated cores of the CPU or to a specific definitions given by the providers, like Amazon's "ECU".
In the cloud servers offers only the clock frequency of the processor may be referenced.
Note that the lone frequency of the processor is not a good indicator of better performances.
An Intel Xeon E5-2695 v3 @ 2.30GHz is twice as performant than an Intel Xeon E5410 @ 2.33GHz.
It's also necessary to know the exact family of the processor used on the cloud server in order to compare it with some others.
Lastly, you have to know whether the core is dedicated or reserved or not.
However, the lone number of vCore assigned is not a good indicator of the power/performances of the cloud server.
You might have 8 vCores assigned on your cloud server and another cloud server from another provider might have better computing resources even with just 1 vCore - all the remaining architecture being the same.
Disk I/O indicates all the INput and OUTput operations between a server and a storage.
Some providers like Stellar give more specific information about disk I/O performances, while other providers don't even mention this factor which is among the most critical to consider.
I/O operations per second.
It is used as the unit of measure to quantify the speed of the disk.
In some cases the number of I/O operations is measured in MB/s instead of IOPS because it's more comprehensible as a unit of measure.
Disk I/O and Disk IOPS are two characteristics that a few providers allow to increment as your growing needs require.
A lot of providers state the type of disks used on their storage and in some cases the type of architecture is referenced, that is, whether they use a network storage or local storage (ie. disks directly connected to the physical Hosts).
Some providers clearly state they use SSD disks in their offers just to look more performant.
Some providers consider as an extra cost the whole IN and OUT traffic of the server, in some other cases the extra cost is for the consumption of bandwidth beyond a certain value.
The amount of CPU, RAM and storage of a virtual server, of a virtual machine (VM), of a Cloud Server or an instance.
It is referenced to with other names, but the main idea is the same.
It's a type of storage that is used as a temporary memory for cloud virtual servers and it's not meant as a data archive.
To speak clearly, it's a disk space used as a Cache, like a RAM memory but on disk.
It controls the incoming and the outgoing traffic blocking all the non authorised connections. It can be implemented as an hardware or as a software solutions.
If the firewall is not mentioned in the cloud server offer, that means that security must be granted on the Operative System level of the cloud server.
In practice, let's say you use Microsoft Windows, you must use the integrated firewall or the one integrated in the antivirus of your choice.
Some providers like Stellar offer virtual server that are linearly scalar, which means that every element of the configuration is independent from the others.
You can increase CPU (in individual steps), RAM (in individual GBs) or the Storage dimension without having to increase the other components.
The majority of providers, like Amazon or Microsoft Azure, offer servers whose configuration characteristics can be modified only by switching to an higher or lower tier.
Changing tier implicates the increase or decrease of several elements, and not just the ones we want to touch.
To make an example, if you have a virtual server with 1 Core, 2GB of RAM and 40GB of Hard Disk and you want to increment the disk size up to 150GB, then you need to buy a virtual server, whose disk size is 150GB, belonging to an higher tier, and perhaps you'll buy even more GB as "just" 150 are not available.
In this example case you might find yourself buying a virtual server with 4 Cores, 16GB of RAM and 200GB of Hard Disk.
It distributes the requests and the work load among the machines to optimise the efficiency.
It can be implemented on both an hardware and a software basis.
Pre-configured Virtual Machines
They are Virtual Machines configured according to the different tiers or levels and there are no customisable elements.
For instance, providers like Amazon EC2 AWS offer plans of instances as preset amount of CPU, RAM and Storage, and resources can be configured singularly.