Storage Class Overview¶
Each object stored in Tebi has a storage class associated with it.
For example, if you list objects in a Tebi bucket, the console shows the storage class for all the objects on the list. Tebi allows you to define your own storage classes in the
Storage Classes section of the client’s console. Define and choose a class depending on your use case scenario, performance access needs, and durability requirements.
Storage class defines where to physically store your data and how many copies you want to store at each location. Storage location makes a huge difference on the access speed of your data – the closer the data to the users, the faster it can be downloaded.
If you want to achieve the best download speed and lowest access latency, we suggest you store at least 1 copy of your data in each region where significant traffic is expected. You can even store multiple copies of the same file in the same location if you expect to have a lot of requests for that particular file. The more copies you store, the more users can access that file simultaneously.
Your files are always accessible in all regions, even if you do not physically store them in all regions. However, access speed will be reduced in these cases, because our system must fetch the data from a remote region. This process is transparent, automatic, and does not cost extra money.
You can always change storage class for any of your files. Doing so allows you to have storage class with many copies of your hot data. You can change to a different storage class with a smaller number of copies to save storage costs.
The minimal number of copies you can define is 2. This is necessary to ensure data redundancy, which would allow us to restore your data in the event of a drive, server, or even datacenter (if you keep each copy in a different datacenter) failure. This process is transparent, automatic, and unnoticeable.