When deciding to use cloud storage to give your business applications the flexibility and scalability they need, your business will be faced with choosing a particular cloud storage system. We suggest that you use a short tutorial to help you decide what type of cloud storage is best for your projects. Initiated to the game also advises on how to combine different types of storage to improve system functionality and save money.
There are three types of cloud storage today: block storage, object storage, and file storage. If we take a look at the characteristics of such storage in the context of real tasks, we will see their advantages and disadvantages.
Option 1: Block cloud storage or disks
Cloud disks mimic the physical disk medium that operating systems typically use. These disks connect to cloud servers and then interact with them just like the disk space of a regular block storage device that serves physical servers. However, the “virtuality” of these disks brings in more functionalities impossible or too expensive to obtain on physical disks.
Features of cloud disks
Cloud drives are flexible to manage. This means that additional disks can be connected to a running cloud server without stopping it and without changing the disk type while it is running. It is possible to create snapshots and disk images and backups. At the same time, you can work with cloud disks in the usual way, as they mimic physical storage devices.
Also, if you are using cloud drives, you don’t have to worry about purchasing and configuring the equipment and other technical nuances of storage operation like you do with physical drives. This will be handled by the cloud provider. It also guarantees a certain system throughput and storage performance – just select the type of disk for your tasks.
Servers, including virtual ones, access block storage through a dedicated data network.
The main types of bulk storage
There are several types of cloud drives, which differ in several ways:
- Hard drives are the lowest performing and cheapest basic drives. They are most often used as the basis for file storage and operating system data downloads.
- SSDs are standard drives, faster than cloud hard drives, but also more expensive. They are used for databases, telemetry data, message queues.
- High IOPS SSDs are fast drives with higher performance and price than standard SSDs. They are mainly used for the same applications (analysis, telemetry, database) but are more loaded. Therefore, they have more serious performance requirements.
- Low Latency NVMe is blazing fast and productive with the lowest latency but also the highest price tag. They are used for high performance databases, analytics, and caching applications where minimal latency is essential.
The choice of block storage type is determined by the type of application and its workloads. We’ll talk more about the selection algorithm below. The price of such storage depends on the type of drive. From hard drive to high IOPS, SSD increases the read and write speeds needed for application performance. Along with low NVMe latency and read / write speed, there is also response speed, which is important for databases and heavily loaded systems with many simultaneous requests and changes. As a result, the cost of different drives increases with performance and response speed: HDD will be cheaper, while low latency NVMe will remain the most expensive.
Option 2. Object Cloud S3 storage
In object storage, S3 data is not stored as files but as objects with metadata. Metadata are labels indicating certain information about an object. For example, if you store video clips in the repository, you can specify the title, author, year released, and other information. In addition, each object has a unique identifier.
To put the data in S3, it is not necessary to bring it to a certain type, so this storage is good for unstructured data, media content and archives.
S3 storage is also ideal for scenarios where a large number of users need to access files at once. For example, video hosting or online movie theaters can place a large number of video clips there. So when millions of visitors access it, S3 can support that load.
Backups and copies of disaster recovery infrastructures are often stored in object storage because it costs less than putting that data in regular storage.
S3 Cloud Storage Features
In object storage, there is no limit to the amount of data stored and it is possible to store any type of object, including unstructured data. If S3 is using multiple applications simultaneously, it does not affect the speed of object distribution: users will get a response to requests quickly. In addition, the storage is automatically scalable. Any amount of data can be placed there; S3 will adapt to the load, even if this volume increases unexpectedly by several factors.
The lifecycle of objects stored in S3 can be flexibly managed. For example, you can configure the deletion of archived data in automatic mode. In this case, unnecessary information will not take up space. And for the automatic processing of objects (such as photos or videos), it is possible to configure a mechanism for notifying Webhooks events. For example, you can set up automatic processing of video after creating an object in the repository.
Access to the files in the repository is possible through URL objects, i.e. the usual link. It allows the user to access the files in the repository without the ability to modify or delete them. A simple example: in an online cinema site, we include a link to a video clip in the object repository. Any visitor can start the video and watch it, but no further actions will be available.
Option 3. File storage
This is where the data is placed in directories and folders with strict hierarchy. It is suitable for shared documents, user files, and large files that need to be changed frequently. On the Mail.ru Cloud Solutions platform, you can connect file storage as a service – use it as remote file storage by connecting it to a cloud server.
Features of file storage
File storage is supported by most conventional systems. This type of storage is therefore good for legacy applications which are difficult or impossible to connect to another cloud storage system. File storage capacity can be decreased and increased, but only in manual mode. Only hard drives can be used for file storage on MCS platform, so its main performance characteristics will be similar to hard drive settings.