The Ceph project is at the core of the "cloud" nowadays. Popular virtual server platforms, such as ours, ColossusCloud, and private cloud applications such as OpenStack, depend on Ceph clusters to host the disks of virtual servers.
What do we mean by that?
Every virtual server needs a disk of course, to store the files of the operating system and any data you upload or generate. To make it simpler, it's basically the C: drive in your Windows computer.
Traditionally, that drive would be a hard disk or SSD connected to the computer running the operating system, but disks being a single point of failure, if the disk fails, your operating system, and its data, are gone. And of course, if the computer itself fails, then you can't use the operating system.
Ceph comes in to solve this problem.
Ceph takes the disk of an operating system, makes three copies of it, breaks it up into a lot of pieces, and distributes those pieces across hundreds of SSD disks across many individual servers. It manages and orchestrates this distribution of data in real time.
In a cloud environment such as ColossusCloud, the disk of your virtual server is stored in Ceph. When your virtual server starts up, it boots the disk and its contents from Ceph, with your virtual server's operating system being completely unaware of Ceph; the operating system sees Ceph as simply a local disk.
It is, truly, quite a remarkable technology.
And in the case a hypervisor fails, the server that runs your virtual server's operating system, we can easily start the virtual server in another hypervisor. Hypervisors and Ceph servers are completely separate and individual pieces of equipment.
At ServerPoint.com, we use Ceph to host the disks of tens of thousands of Linux and Windows virtual servers, providing a level of reliability and uptime that can't be matched by other non-Ceph solutions.