Storage Virtualization - Unveiling the magic that happens in cloud storage!

 Okay....you have created a virtual machine and in the process, it might have asked you about something ... something about virtual hard disk...
Ah... yeah....that's the way virtual machines store stuff and things inside. But really beneath the ground, what happens?

Yeah...let's dive in and learn about "Storage Virtualization"!

You might ask me, "Hey, are they storing stuff up in the air coz' you're saying (quoting in the air) virtualized ? "

Actually, still virtual machines running in the cloud servers do use physical storage media such as hard drives, solid state drives (SSDs), etc. to store but the complex I/O operation is just abstracted away!

Wait, there are some types of storage virtualization ... let's see what they are!

1) Block level 

Here, similar to our computers, servers running the virtual machines just write stuff directly onto your physical storage media.

2) Object-level

In here, data isn’t stored on your disk directly. Instead, it’s abstracted away into data buckets. This data is accessed by API calls from your application. For large amounts of data, this can be a more scalable solution than block storage. This means you don’t have to worry about running out of space after setting up your buckets.

3) File-Level

People use file server software like Samba and NFS when they want another server to host their files. The files are saved in folders called shares. This removes the need to manage disk space and also allows multiple people to share a storage device. Servers, virtual servers, and desktop computers all can take advantage of file servers.

4) Host-based

With host-based storage virtualization, accessing the host or devices attached to the host is allowed. A driver is installed on the server which intercepts and redirects the IO (Input and Output) requests. Typically, these IO requests go directly to a hard drive but they could also be routed to other devices such as a USB flash drive. The most common use for this type of storage is accessing physical installation CDs or DVDs so that an operating system can be easily installed on the virtual machine.

5) Network-Based

A fiber channel switch is placed between the host and the storage. The switch is where the virtualization occurs and redirects the I/O requests. This method works with any operating system without special drivers.

6)Array-Based

A master array handles all of the IO requests for all of the arrays. This allows for management from a central location and simplifies data migrations.

That's all folks! I hope you might've understood about storage virtualization.

Until next time, never stop learning!

Comments

Popular posts from this blog

Mounting an EBS volume to an EC2 Linux instance