UNIT II: Cloud Storage and Data Management

PART A: FULL NOTES (For Learning & Long Answers)

1. Introduction to Data Storage

Data storage is the process of saving digital data in a storage medium for future access. In cloud
computing, data storage is provided as a service by third-party cloud providers over the internet. It
removes the need for organizations to invest in physical storage infrastructure.

Data storage plays a vital role in cloud computing as all cloud services depend on data availability,
durability, and security. Cloud storage supports backup, disaster recovery, data sharing, and long-
term archiving.

Key Features:

 Remote storage in cloud data centers

 Internet-based access

 High scalability and availability

 Fault tolerance using replication

2. Storage Devices

Storage devices are hardware or virtual components used to store data. They are classified into block
storage and file storage devices.

2.1 Block Storage Devices

Block storage stores data in fixed-sized blocks. Each block has a unique address and can be managed
independently by the operating system.

Characteristics:

 High performance and low latency

 OS manages the file system

 Used for databases and virtual machines

Examples: HDD, SSD, SAN

2.2 File Storage Devices

File storage organizes data into files and directories and maintains its own file system.

Characteristics:

 File-level access
  Supports shared access

 Commonly implemented using NAS

3. Cloud Storage Classes

3.1 Unmanaged Cloud Storage

In unmanaged cloud storage, storage is preconfigured by the cloud provider. Users cannot format or
change file systems.

Advantages: Simple to use, low management overhead

Limitations: Less flexibility and control

3.2 Managed Cloud Storage

Managed cloud storage provides raw storage to users, allowing them to partition and format as
required.

Advantages: High flexibility, scalability, and control

4. Creating a Cloud Storage System

Cloud storage systems are built using virtualization and data replication techniques.

 Data is replicated across multiple servers and locations

 Virtualization aggregates storage resources

 StorageGRID provides centralized management

 CIFS and NFS enable file access

5. Virtual Storage Container

A virtual storage container is a logical unit that groups storage resources such as files, blocks, and
objects.

Features:

 High performance

 Uses Logical Unit Numbers (LUNs)

 Supports multiple storage formats

Challenges:

 Data security and privacy

 Data location transparency

6. Enterprise Data Storage

Enterprise storage is a centralized repository for storing business-critical data.

Characteristics:

 High availability and reliability

 Centralized management

 Designed for heavy workloads

Types:

 Direct Attached Storage (DAS)

 Network Attached Storage (NAS)

 Storage Area Network (SAN)

7. Direct Attached Storage (DAS)

DAS is directly connected to a server or computer without using a network.

Features:

 Block-level access

 Uses Host Bus Adapter (HBA)

 Low cost and simple architecture

Types: Internal DAS, External DAS

8. Network Attached Storage (NAS)

NAS is a file-based storage system connected to a network.

Protocols: NFS, SMB, AFP

Advantages: Easy setup, centralized file sharing, scalable

Disadvantages: Slower than DAS, network dependency

9. Storage Area Network (SAN)

SAN is a dedicated high-speed network that provides block-level storage to multiple servers.

Protocols: FCP, iSCSI, FCoE, NVMe

Advantages: High scalability, security, efficient storage utilization

Data storage management ensures data availability, reliability, and security.

Functions: Backup, recovery, monitoring, capacity planning

11. File System in Cloud

Cloud file systems provide shared file access using traditional protocols.

Benefits: Simplifies application migration, supports collaboration

12. Cloud Data Stores

Cloud data stores store digital data on remote servers managed by providers.

Examples: Google Drive, Dropbox, OneDrive

13. Cloud vs Grid Computing

Cloud computing is service-oriented with centralized management, whereas grid computing is

14. Cloud Data Management

Cloud data management handles storage, governance, security, and lifecycle of data.

Benefits: Security, scalability, automated backups

15. Cloud Provisioning

Cloud provisioning is the allocation of cloud resources to users.

Types: Network, Server, User, Service provisioning

16. Data Intensive Technology

Data-intensive computing processes large volumes of data efficiently.

Characteristics: Scalability, reliability, minimal data movement

17. Cloud Storage from LANs to WANs

Characteristics: Elastic computing, data replication, remote storage

Distributed storage spreads data across multiple nodes.

Technologies: Amazon Dynamo, CouchDB, ThruDB

PART B: REVISION NOTES (For Quick Exam Preparation)

Key Definitions

 Cloud Storage: Storage of data on remote servers accessed via internet

 DAS: Storage directly attached to server

 NAS: Network-based file storage

 SAN: High-speed block storage network

Storage Types (One Line)

 Block Storage: Raw storage blocks

 File Storage: File and directory-based storage

Cloud Storage Classes

 Unmanaged: Preconfigured, less control

 Managed: User-controlled, scalable

DAS / NAS / SAN (Quick View)

 DAS: Simple, low cost, no sharing

 NAS: File sharing, network-based

 SAN: High performance, scalable

Cloud vs Grid (One Line)

 Cloud: Service-oriented, centralized

 Grid: Application-oriented, decentralized

Cloud Provisioning Benefits

 Speed

 Cost savings

Data Intensive Computing

 Processes Big Data

 Focuses on scalability and reliability

Distributed Storage Examples

 Amazon Dynamo

 CouchDB

 ThruDB

END OF UNIT II – FULL NOTES + REVISION NOTES