Storage Fundamentals

Module 1: Introduction to Data Storage

TABLE OF CONTENTS
INTRODUCTION TO DATA STORAGE .................................................................................................... 5
MODULE OBJECTIVES .......................................................................................................................... 5
SINGLE USER - HARD DISK DRIVE STORAGE ......................................................................................... 5
MULTIPLE USERS - HARD DISK DRIVE STORAGE ................................................................................... 6
CENTRALIZED SERVER STORAGE........................................................................................................... 6
DIRECT ATTACHED STORAGE (DAS) ..................................................................................................... 6
DISADVANTAGES OF A DAS ENVIRONMENT ........................................................................................ 7
NETWORKED STORAGE ....................................................................................................................... 7
NETWORKED STORAGE ....................................................................................................................... 7
ADVANTAGES AND LIMITATIONS OF STORAGE TECHNOLOGIES .......................................................... 8
ADVANTAGES AND LIMITATIONS OF NETWORKED STORAGE TECHNOLOGIES ..................................... 8
LEARNING ACTIVITY INTRO ................................................................................................................. 8
MODULE SUMMARY ........................................................................................................................... 9
MODULE 1: INTRODUCTION TO DATA STORAGE

INTRODUCTION TO DATA STORAGE

In a small organization, data storage is usually simple. As the organization grows, the storage needs also
grow. This module introduces the various storage architectures that are used to store data.
The approximate duration for this module is 25 minutes.

SINGLE USER - HARD DISK DRIVE STORAGE

Both individual users and businesses have data storage needs. However, individual data storage needs
may not increase as much as the storage needs of a business. Let us look at an example to see how
storage needs may increase as the business grows.
Tom is the owner of a small grocery store. He runs the entire business using a single desktop computer
using applications such as Microsoft Word, Excel, PowerPoint and Access.
Apart from the operating system and applications, he stores all the user files, such as inventory details
and transactions, on the internal magnetic disk called the hard disk drive of the computer.
When Tom wants to retrieve any file, for example the inventory details, he simply finds the file and double-
clicks on it. The computer retrieves the associated application and the file from the hard disk and loads
them into the Random Access Memory (RAM) and Tom can then view the file on the screen.
The type of storage that Tom is using is a single hard disk drive.
The computers internal hard disk drive provides the storage for the business. As prudent business owner,
Tom takes a backup at regular intervals to avoid data loss.

MULTIPLE USERS - HARD DISK DRIVE STORAGE

As Toms business grows, he decides to increase the number of checkout counters and hires people to
man them. Each counter has a computer and there are separate back office computers to maintain
inventory and accounts. Each of these computers has a separate hard disk drive where the relevant data
is stored.
Tom decides to connect the computers in a peer-to-peer network so that he can access sales and
inventory details from any computer in the store.
In a peer-to-peer network, though the computers are connected, data is still stored in the hard disk drives
of individual computers. One common problem in this set up is file versioning. For example, an employee
may copy the inventory file from another computer and update the local copy. The employee may then
forget to update the master copy.
Apart from file versioning problems, Tom finds that he is spending a lot of time backing up data on
individual machines.

CENTRALIZED SERVER STORAGE

To resolve these issues, Tom consults his neighbor Jason, who is a network administrator at an IT firm.
Jason advises Tom to buy a server. He explains that a server is a dedicated system with a large internal
hard drive, or multiple hard drives. It runs on a server operating system.
The centralized server may store shared applications. The servers hard drive provides a centralized
storage that employees can access over a client-server network.Instead of data being stored on individual
hard disk drives, all important data is now at a centralized location. For example, the server will store the
inventory details and when any employee wants to update the inventory, they can access it from the
centralized server and save the updated file in the server. This will help reduce versioning problems. Each
employee may also be provided storage space in the centralized server so that they can access their files
from any computer in the office. Backup is also easier since only the server has to be backed up.

DIRECT ATTACHED STORAGE (DAS)

By installing a server, Tom has moved from using single hard disks to a form of storage called Direct
Attached Storage (DAS). A DAS system consists of simple and affordable storage devices that are ideal
for localized file sharing in an environment with a single server or multiple servers.
Let us look at Toms store again. As the neighborhood grows, Toms business expands too. He hires a
network administrator, Richard. Richard notices that the hard disk is getting full because of all the
additional data. Instead of adding additional hard drives to the server, he decides to implement multiple
servers. He adds a couple of file servers, moves to a SQL Server database, and adds a web server to
enable customers to order online. He also adds a mail server to manage mail services. He connects the
servers in a Local Area Network (LAN).

DISADVANTAGES OF A DAS ENVIRONMENT

At this stage, though there are multiple servers, Richard is still using DAS storage. Each server comes
with its own storage.
As his business grows, Tom opens a chain of retail stores across the state. The amount of data that the
business generates is now huge. In the first couple of stores, Richard uses the same DAS storage model.
However, he immediately runs into problems.
One problem is that of storage utilization. Richard notices that the file server storing the daily transactions
is utilized more than the server providing mail services. He will need to add another hard disk drive to the
file server, even though the mail server has unused storage. Another problem he faces is that of data loss.
A file server with a large amount of data suddenly crashes. The back up for that file server has not been
taken and so a large amount of data is lost. Richard finds that taking a back up of multiple servers
involves a lot of time and effort.

NETWORKED STORAGE
To address these problems, Richard now informs Tom that it will be better to use networked storage. He
explains that in a networked storage environment, the servers are separated from the storage so that
many servers can access a common pool of shared storage. Servers or storage can be added
independently. Users can access the storage either directly or through the servers. Networked storage
can also be accessed in different methods. Small amounts of data, called blocks, such as records from a
database, or entire files, such as documents or spreadsheets can be accessed. This increases flexibility
and reliability.
In the case of Toms store, Richard decides to have all the storage in a single location. As new stores are
opened, the amount of data increases. If the storage is full, he can add additional storage without
disturbing the servers.
Richard explains to Tom that there are two types of networked storage, a Storage Area Network (SAN)
and

Network Attached Storage (NAS).

A SAN is a networked storage system that interconnects different kinds of data storage devices with
associated data servers. A SAN typically is a network of storage devices. A SAN provides servers access
to the storage devices using the Fibre Channel standard. It provides block-level data transfer.
NAS is a network storage system. In NAS, a storage server is attached to a companys network using file-
level protocols such as, Network File System (NFS) for UNIX environments and Common Internet File
System (CIFS) for Windows environments. NAS offers file-level data sharing across the network.
Richard informs Tom that he can opt for a SAN system if he wants better disk utilization. But the Fibre
Channel technology is expensive. If he does not want to invest much in a storage network, he can opt for
a NAS system. However a NAS is not as effective as SAN in transferring large blocks of data.
Richard explains that there is another type of storage system that is like a SAN system but not as
expensive. It is called an Internet Protocol Storage Area Network (IP SAN). In this type of storage, an
alternative technology called iSCSI is used instead of Fibre Channel.

ADVANTAGES AND LIMITATIONS OF STORAGE TECHNOLOGIES

We have discussed several storage technologies. Let us quickly take a closer look at the advantages and
limitations of each of these technologies.
The single hard disk drive storage is the most common form of data storage. It is portable and has high
read/write speed. Backing up data from a single hard disk drive is easy. Some of the limitations of single
hard disk drive storage are that it has a limited capacity to store data and without additional support, it
cannot be shared on multiple computers.
The advantage of using DAS is that it is easy to deploy and manage. A storage device is attached to the
server and if that storage is fully utilized, additional storage can be attached. For small organizations that
do not transfer files over long distances, DAS is a cost-effective storage solution. The main disadvantage
of a DAS storage system is that it is inconvenient for data transfer in a network. Since all the systems can
access the storage space only through the server, the processing burden on the server is increased.
Since each server has its own, separate storage, backup becomes more difficult and time-consuming as
the environment grows.

ADVANTAGES AND LIMITATIONS OF NETWORKED STORAGE TECHNOLOGIES

In a SAN system, the disk utilization is high because all the hard drives are connected in a network. Since
the storage processing is moved to a separate network, application performance is better. The data is
stored centrally in a SAN environment and so it is easier to manage and backup. The major limitation of a
SAN system is that it is very expensive and difficult to deploy and manage.
Since NAS is a dedicated storage device connected to a network, it provides high storage capacity and
fast access for multiple clients. Since NAS is part of a network, it is easy to share data among the clients.
When there are large blocks of data to be shared, it is easier to use a SAN than a NAS.

LEARNING ACTIVITY INTRO

Module 2: Networked Storage

TABLE OF CONTENTS
Networked Storage ............................................................................................................................. 6
Module objectives .............................................................................................................................. 6
Storage Architecture and Technologies ............................................................................................... 6
Lesson Objectives ............................................................................................................................... 6
NAS System ....................................................................................................................................... 7
How a NAS System Works? ................................................................................................................. 7
Components of a SAN System ............................................................................................................. 7
How a SAN System Works ................................................................................................................... 8
Components of an IP SAN System ....................................................................................................... 8
How an IP SAN Works ......................................................................................................................... 8
Learning Activity ............................................................................................................................... 9
Lesson summary................................................................................................................................ 9
Storage Strategies ............................................................................................................................... 9
Lesson Objectives .............................................................................................................................. 10
Scenario .......................................................................................................................................... 10
Scenario .......................................................................................................................................... 10
Lesson summary.............................................................................................................................. 11
Market Trends ................................................................................................................................ 11
Lesson Objectives .............................................................................................................................. 11
Criteria for a Good Storage Solution - Application ............................................................................. 11
Criteria - Data Protection .................................................................................................................. 12

Criteria - Data Availability ................................................................................................................. 12

Criteria - Data Security ...................................................................................................................... 12
Criteria - Performance and Scalability ................................................................................................ 13
Criteria - Cost .................................................................................................................................. 13
Government Regulations - HIPAA and GLBA ...................................................................................... 13
Government Regulations - SOX and SEC Rule 17a .............................................................................. 14
Technology Trends - Tiered Storage and NAS and SAN Gateways ...................................................... 14
Technology Trends - Storage Virtualization and SATA ........................................................................ 15
Technology Trends - Cloud Storage .................................................................................................... 15
Learning activity intro........................................................................................................................ 15
Lesson summary.............................................................................................................................. 16
Module summary .............................................................................................................................. 16
NETWORKED STORAGE

Organizations use networked storage methods such as, NAS, SAN, and IP SAN to store large amounts of
data. The architecture that is used by an organization depends on various parameters such as, cost,
reliability, and availability. This module explains the networked storage methods and the parameters that
have to be considered while planning a storage solution.

STORAGE ARCHITECTURE AND TECHNOLOGIES

In organizations where storage needs are high and where data storage has to be independent of the
application servers, networked storage is the ideal solution. The type of networked storage system to be
installed depends on the infrastructure need of the organization and the funds available for network
storage.

NAS SYSTEM

Network Attached Storage (NAS) is a backup and storage system that is connected to a network. NAS is
used within organizations that have several departments and have large data storage requirements.
A NAS environment has a hardware device called the NAS box or a NAS head that acts as the interface
between the storage system and the network clients.
Clients access the NAS over an Ethernet connection. On a network, the NAS system appears as a node
with the IP address of the NAS head.
Clients communicate with the NAS head using TCP/IP protocols such as NFS and CIFS. Many NAS
systems also support web access through HTTP.

HOW A NAS SYSTEM WORKS?

Network clients can access the storage devices through the NAS head. The NAS head can be a
processor that runs its own embedded operating system. The NAS head implements file services by
using protocols such as NFS and CIFS. In a NAS environment the application servers handle all the
processing of data but the data is stored and retrieved from the NAS system. The NAS system works as a
self-contained computer that is connected to a network, and it supplies the file-based data storage
services to devices on a network.
The NAS head consists of the processor and an array of hard disks using RAID technology to protect
against disk failure. For additional storage, external disk drives can be attached to the NAS device.

COMPONENTS OF A SAN SYSTEM

A Storage Area Network or SAN is a network of storage devices. It connects storage and application
servers through a network known as a "fabric". This network is built with hardware called switches.
There are three major components in a SAN environment. They are the servers, the SAN infrastructure or
"fabric", and the storage systems. Servers providing many different applications and services running
many different operating systems can all participate in the same SAN environment.
The SAN infrastructure includes the hardware, cabling, and software components that help data move
into and within the SAN. The management software helps to monitor the entire network and configure
individual components. In a SAN environment, the storage system is connected to the fabric and is not
directly connected to any server bus. The most commonly used protocol in a SAN environment is the
Fibre Channel Protocol (FCP). It is a high-speed network technology primarily used for storage
networking. iSCSI is another commonly used protocol. It transports storage requests over standard
TCP/IP networks.

Another technology called Fibre Channel over Ethernet or FCoE is gaining importance and popularity. It
allows Fibre Channel and TCP/IP to be run over a single, "converged" physical network.

HOW A SAN SYSTEM WORKS

In a Fibre Channel SAN system, each server will have a Host Bus Adapter or HBA installed.
Each HBA connects to the Fibre Channel switches in the SAN infrastructure. These switches are
connected to the storage network. The switches route traffic from the servers to the storage network. The
switches can detect failed connections and reroute data to the correct device. Since a SAN system
performs block transfer, it addresses the data by numbers called logical unit numbers or LUNs.
The storage network is separate from the user network, providing higher security and better performance,
but also increased administrative overhead and cost.

COMPONENTS OF AN IP SAN SYSTEM

An IP SAN is a type of SAN environment that uses iSCSI instead of Fibre Channel for data transmission.
You can use IP SAN in applications where reliability, flexibility, and scalability are required, but cost is a
factor. Fibre Channel SANs require a Fibre Channel switch infrastructure to be installed and all hosts
require Fibre Channel HBAs. This can prove to be costly for small organizations. Such organizations can
use an IP SAN to manage their storage.
An IP SAN uses the iSCSI protocol for transmitting data access requests over an Ethernet connection.
Using standard Ethernet cables, adapters, and switches, it is possible to build a low-cost IP SAN.
The main components for an IP SAN system are the iSCSI initiators and iSCSI targets. An iSCSI initiator
refers to a server that initiates iSCSI command sequences. An iSCSI target refers to the storage devices
that are iSCSI enabled.
The servers in an IP SAN system will have an iSCSI initiator installed. The servers connect to the iSCSI
targets using the Ethernet network.

HOW AN IP SAN WORKS

In the IP SAN environment, all the storage is available to all the servers. The servers will have an iSCSI
initiator installed. The iSCSI initiator can be a hardware device similar to a host bus adapter or it can be
software. Software iSCSI initiators are included with almost all operating systems such as Windows and
UNIX. These software iSCSI initiators use standard Ethernet ports to connect to the iSCSI target.
After the iSCSI initiator has been installed, it is used to log in and establish a connection with the storage
device. After the iSCSI initiator has logged in, the storage device appears as a local hard disk to the
operating system. The iSCSI initiator then sends iSCSI commands and the storage device interprets the
commands and processes them.

LEARNING ACTIVITY
LESSON SUMMARY
In this lesson, you learned that:
The main components of a NAS storage environment are the NAS head, the Ethernet network, and the
protocols
The most commonly used protocols in NAS are TCP/IP, NFS, and CIFS
The application servers connect to the storage devices through the NAS head
The NAS system performs file level data transmission
The main components of a SAN environment are servers, storage devices, and switches
The most commonly used protocols in a SAN environment are FCP, iSCSI, and FCoE
The SAN system performs block-level data transmission
The main components of an IP SAN are iSCSI initiators, iSCSI targets, servers, and the Ethernet
switches
The servers will have an iSCSI initiator installed

STORAGE STRATEGIES

There are different ways in which data can be stored in an organization. An organization can choose to
use DAS, SAN, NAS, or IP SAN based on its requirements and available infrastructure.