Xml and More: OCI―Knowing Block Volume Service Basics

Video 1. OCI Level 100 - Block Volume (Demo starts at 27:00)

Oracle Cloud Infrastructure (OCI ) offers two main storage services:

Block Volume Service

Integrates with IAM for authentication and authorization, for all interfaces (the Console, SDK or CLI, and REST API)
Block storage operates at the raw storage device level and manages data as a set of numbered, fixed-size blocks using protocols such as iSCSI
Block Volume Service lets you dynamically provision and manage block storage volumes
Block Volume uses the Advanced Encryption Standard (AES) algorithm with 256 bit key for encryption

Data is encrypted at rest in both volumes and backups

You can create, attach, connect, and move volumes, as needed, to meet your storage and application requirements

Once attached and connected to an instance, you can use a volume like a regular hard drive
Volumes can also be disconnected and attached to another instance without the loss of data
Volumes are only accessible to instances in the same Availability Domain (AD)

You cannot move a volume between availability domains or regions
However, block volume backups can be restored as new volumes to any availability domain within the same region they are stored

Object Storage Service

See [1] for more details

In this article, we will cover Block Volume Service. But, before we start, we will begin with a related OCI Storage Option―local NVMe SSD.

Figure 1. OCI Storage

Local NVMe SSD Storage

OCI provides a variety of instance configurations in both bare metal (BM) and virtual machine (VM) shapes. Each shape varies on multiple dimensions including memory, CPU cores, network bandwidth, and the option of local NVMe SSD storage found in Dense IO shapes.

NVMe SSD Devices are locally attached to some instance shapes in OCI. It has the following features:

It can be used for workloads (e.g., Big Data, OLTP) requiring high storage performance
Data are persisted during reboot or pause
These devices are not protected

Customers are responsible to protect and manage the durability the data on these devices
A protected RAID array is the recommended way to protect against an NVMe device failure. There are three RAID levels that can be used for the majority of workloads:

RAID 1

Blocks mirrored

RAID 10

Blocks mirrored and striped

RAID 6

Blocks striped with dual parity across drives

OCI provides a service-level agreement (SLA) for NVMe performance

Block Volume Service

A common usage of Block Volume Service is to add storage capacity to an instance. Block volumes provide you persistent and durable storage. All volumes are automatically replicated for you, helping to protect against data loss. Any data will remain intact until you reformat or delete the volume. There are two types of volumes:

Block volume

A detachable block storage device for your data storage

Boot Volumes

Detachable system disks for your Compute instances
When you terminate an instance, you can keep the associated boot volume and use it to launch a new instance using a different instance type or shape

To use a block storage volume, you need to follow below four steps:

Create a block storage volume through the console (see Figure 3) or the API
Attach the volume to an instance using a volume attachment (See Figure 2)
Connect to the volume from your instance's guest OS using iSCSI
Mount the volume and use within your instance

Figure 2. Attach Block Volume

Backup Policy (Options in Step 1)

The OCI Block Volume service provides you with the capability to perform volume backups automatically on a schedule and retain them based on the selected backup policy. Each backup policy has a set backup frequency and retention period. This allows you to adhere to your data compliance and regulatory requirements. See Overview of Block Volume Backups for more information.

When you create a new volume on OCI, you can select the appropriate backup policy at that time:

None

If your requirements change

You can adjust the schedule and retention period by selecting a different backup policy
You can also do manual backups

Bronze

Monthly incremental backups

Run on the first day of the month. Incremental backups are retained for 12 months

Yearly full backup

Run yearly on January 1st. Full backups are retained for 5 years

Silver

Weakly incremental backups

Run on Sunday. Incremental backups are retained for 4 weeks

Monthly full backup

Run on the first day of the month and are retained for 12 months

Yearly full backup

Run yearly on January 1st. Full backups are retained for 5 years

Gold

Daily incremental backups

Incremental backups are retained for 7 days

Weakly incremental backups

Run on Sunday. Incremental backups are retained for 5 weeks

Monthly full backup

Run on the first day of the month and are retained for 12 months

Yearly full backup

Run yearly on January 1st. Full backups are retained for 5 years

iSCSI vs Paravirtualized (Options in Step 2)

At step 2, you can choose how to attach your block volume in two ways:

iSCSI

A TCP/IP-based standard used for communication between a volume and attached instance
Are the only option when connecting block volumes to

Bare metal instances
VM instances based on Windows images published prior to February 2018
VM instances based on Linux images published prior to December 2017.

Paravirtualized

A virtualized attachment available for VMs
Once you attach a volume using the paravirtualized attachment type, it is ready to use, you do not need to run any additional commands.
However, due to the overhead of virtualization, this reduces the maximum IOPS performance for larger block volumes

Note that you should only attach Linux volumes to Linux instances and Windows volumes to Windows instances.

Step 3 and 4

You connect (step 3) or mount (step 4) block volumes differently based on below considerations:

Volume Attachment Type

Paravirtualized

The volumes are connected automatically

iSCSI

You still need to connect and mount the volume from the instance for the volume to be usable

For more information, see Volume Attachment Types and Connecting to a Volume

OS Type

Linux-Based Instances

On Linux operating systems, the order in which volumes are attached is non-deterministic, so it can change with each reboot

If you refer to a volume using the device name, such as /dev/sdb, and you have more than one non-root volume, you can't guarantee that the volume you intend to mount for a specific device name will be the volume mounted

If you want to mount automatically volumes on instance boot, you need to set some specific options in the /etc/fstab file, or the instance may fail to launch. This applies to both iSCSI and paravirtualized attachment types.
Consistent Device Path