Your choice of disk will determine how redundant your data is and how fast you can access your data.
Google Compute offers three broad types of disk. The first of which is the standard persistent disks (HDD). These disks give you flexibility, in terms of partition sizing and can be attached to any machine type. These disks are encrypted and allow snapshotting, but, are the slowest disks that Google Compute offer (with a max read speed of 0.12mb/second).
Compare that speed the next disk provided by Google – solid state persistent disks (SSD) and you’ll see an immediate four-fold read speed improvement. These disks are extremely low latency and also offer flexible partition sizes.Â These are best used when your limiting factor is random IOPS* or streaming throughput.
*Random IOPS means that your files are scattered all over the drive, not in neat rows or groups, so it takes more work to find. Random IO is the most difficult and time consuming type a storage device must deal with (definition).
The two options above, offer data redundancy, the below, do not.
The local SSD drives are extremely efficient at handling high rates of IOPS, they’re extremely low latency, but are less flexible than those outlined above, as you have a fixed partition size. Local SSDs are particularly useful when you need a fast scratch disk** or cache (and don’t want to use pure RAM).
**A scratch disk is dedicated to temporary storage. It’s not used to backup permenant files. Typically, these type of disks are set to erase data at regular intervals, leaving the disk free for future use.
This article was bought to you by Netshock. Netshock aim to provide technology guides and insight to our readers.