Cloud Storage Types: Object, File, and Block

The Three Pillars of Cloud Storage

Cloud providers offer three main types of storage, each designed for different purposes: Block, Object, and File storage. The choice of storage type directly impacts performance, cost, and how applications interact with the data.

To get a comprehensive overview of all three, let's start with a video that explains each one.

Now, let's break down each type with the level of detail required for your exams.

1. File Storage: The Hierarchical Model

File storage, also known as file-level storage, is likely the model you are most familiar with. It organizes data in a hierarchical structure of files, folders, and directories.

• Definition: A storage methodology where data is stored as a single piece of information inside a file, which is then placed in a folder within a hierarchy of directories. To access the data, a system needs the path to the file (e.g., /home/user/documents/report.pdf).
• Analogy: Think of a physical filing cabinet. Documents (files) are placed inside folders, which are organized within drawers (directories).

Key Characteristics:

• Structure: Hierarchical (tree-like).
• Access: Via file paths using standard protocols.
• Protocols: Network File System (NFS) for Linux/Unix systems and Server Message Block (SMB/CIFS) for Windows systems.
• Pros: Intuitive, easy to use, ideal for shared access among multiple users.
• Cons: Performance can degrade as the number of files and directories grows, making it less suitable for massive scalability compared to object storage.

Use Cases and Examples:
File storage is ideal for collaborative environments and applications that require a shared file system.

• Common Use Cases:
  • Shared corporate drives and home directories.
  • Web content management systems (CMS).
  • Development environments.
  • User-generated content platforms.
• Cloud Service Examples:
  • AWS: Amazon Elastic File System (EFS)
  • Azure: Azure Files
  • GCP: Google Cloud Filestore
  • OpenStack: Manila (Shared File System Service)

2. Block Storage: The High-Performance Model

Block storage is a lower-level storage model that provides raw storage volumes for high-performance applications.

• Definition: A storage technology that breaks data into fixed-size chunks called "blocks." Each block has a unique address but contains no additional information or metadata to provide context. The operating system running on the server is responsible for managing these blocks and organizing them into a file system.
• Analogy: Think of a raw, unformatted hard drive (a Storage Area Network or SAN). The drive provides raw space, and it's up to you to install an OS that formats it and creates a file system (like NTFS or ext4) to store files.

Key Characteristics:

• Structure: Raw blocks of data.
• Access: Mounted as a local disk to a single virtual machine. Provides direct access via protocols like iSCSI.
• Performance: Offers very low latency and high I/O operations per second (IOPS), making it ideal for performance-sensitive workloads.
• Pros: Excellent performance, high efficiency for transactional operations, allows the OS full control over the file system.
• Cons: Can be more expensive, lacks rich metadata at the storage level, and is typically tied to a single compute instance.

Use Cases and Examples:
Block storage is the backbone for applications that require fast, reliable, and granular access to data.

• Common Use Cases:
  • Transactional databases (SQL, NoSQL).
  • Email servers (e.g., Microsoft Exchange).
  • Virtual Machine File Systems (VMFS), used by hypervisors like VMware ESXi.
  • RAID (Redundant Array of Independent Disks) arrays.
• Cloud Service Examples:
  • AWS: Amazon Elastic Block Store (EBS)
  • Azure: Azure Disk Storage
  • GCP: Google Persistent Disk
  • OpenStack: Cinder (Block Storage Service)

3. Object Storage: The Cloud-Native Scalable Model

Object storage is a modern storage architecture designed for storing massive quantities of unstructured data. It is the dominant storage type for the cloud era.

• Definition: A storage methodology where data is managed as "objects." Each object consists of three components: the data itself (e.g., a photo or video file), extensive and customizable metadata, and a globally unique identifier. Objects are stored in a flat address space, called a bucket or container.
• Analogy: Think of a valet car service. You hand over your car (data), and you get a claim ticket (the unique ID). The valet service parks your car wherever is most efficient. You don't know or care about the exact parking spot; you just use your ticket to retrieve your car. The rich metadata is like the information about your car: make, model, color, and when you parked it.

Key Characteristics:

• Structure: Flat (no directories or folders).
• Access: Via a unique ID using HTTP-based REST APIs (e.g., GET, POST, PUT, DELETE commands).
• Scalability: Virtually limitless. It can scale to exabytes of data.
• Pros: Massively scalable, highly durable and resilient, cost-effective for large volumes of data, and allows for rich, searchable metadata.
• Cons: Higher latency compared to block storage, not suitable for transactional data, and files must be modified by overwriting the entire object.

Use Cases and Examples:
Object storage is built for unstructured data that needs to be stored and accessed at scale. The Gate Smashers video mentions "blob storage" in Azure, which is their term for object storage.

• Common Use Cases:
  • Storing and distributing static assets for websites and applications (images, videos, documents).
  • Backup, archiving, and disaster recovery.
  • Big Data analytics, data lakes, and AI/ML model training data.
  • IoT data collection.
• Cloud Service Examples:
  • AWS: Amazon Simple Storage Service (S3)
  • Azure: Azure Blob Storage
  • GCP: Google Cloud Storage
  • OpenStack: Swift (Object Storage Service)

Comparison: Object vs. Block vs. File Storage

For your exam, being able to directly compare these three models is crucial. To start, read through the excellent summary and comparison table in the following article.

To consolidate this information for easy revision, here are the two most important comparison formats for your exams: a detailed table and a visual summary diagram.

Detailed Comparison Table

This table, adapted from the article you just read, is perfect for memorizing the key distinctions.

Visual Summary

This diagram provides a quick, visual summary of the core concepts, advantages, and disadvantages of each storage type. Formats like this are excellent for last-minute revision.

Conclusion

In this lesson, we have dissected the three fundamental types of cloud storage. A clear understanding of when to use Block, File, or Object storage is a hallmark of a proficient cloud practitioner and a key topic for your examination.

Key Takeaways:

• File Storage is for shared, hierarchical data, accessed via a path (NFS/SMB). Think of a network drive.
• Block Storage is for high-performance, transactional workloads, accessed as a raw disk (iSCSI). Think of a high-speed SAN for your database.
• Object Storage is for massive-scale, unstructured data, accessed via an API (HTTP/REST). Think of a limitless digital warehouse for backups, media, and big data.
• The choice always depends on the application's requirements for performance, scalability, data structure, and cost.

Preview of the Next Lesson:

Now that we understand the types of storage, our next lesson will delve into cloud storage architecture. We will examine how cloud providers build these storage systems to be resilient and reliable, covering key concepts like synchronous and asynchronous data replication and data consistency models (strong vs. eventual consistency).