© 2005 emc corporation. all rights reserved. the complexity of information management 第一章...

© 2005 EMC Corporation. All rights reserved.

The Complexity of Information Management The Complexity of Information Management

第一章信息管理的复杂性

© 2005 EMC Corporation. All rights reserved. Storage Systems Architecture - Introduction - 2

本章目标及内容信息，对现代企业管理至关重要，信息不仅仅能够创

造新的商业机会，同时，充分利用信息也能为企业带来市场中的竞争优势。本章从信息的重要性开始讲起，介绍在海量信息管理需求下的存储解决方案与数据中心的基础构架，从而建立起存储的基本概念。

本章内容包括三个部分：1.1 满足现代需求的数据存储（ Meeting Today’s Data Storage Needs ）1.2 数据存储的解决方案（ Data Storage Solutions ）1.3 数据中心的基础构架（ Data Center Infrastructure ）


Meeting Today’s Data Storage Needs Meeting Today’s Data Storage Needs

Module 1.1


Meeting Today’s Data Storage Needs

Upon completion of this module, you will be able to:

Describe who is creating data and the amount of data being created

Describe the value of data to business

Describe the challenges in data storage and data management

List the solutions available for data storage


Data Creation

Data is being created at an ever increasing rate– Data creation/generation is growing at a rate in excess of 50% year-

over-year

– The need to store the data over longer periods of time with improved accessibility is also growing

Information Technology (IT) budgets are responding– IT budgets typically have to account for expenditure on Servers,

Networks, Storage, Personnel, etc.

– To keep up with data storage needs, IT expenditure on Storage has increased proportionally

– It is estimated that about 40% of the IT expenditure is Data Storage related


Data Creation: Individuals

What data is created by Individuals?

Examples include:– Photos

– Documents

– Spreadsheets

– Video


Data Creation: Individuals

Where is this data stored?

Storage can be in:– Cameras

– MP3 players

– Laptop hard drives

– CDROM/DVDs

– USB drives


Data Creation: Business

What data is created by a business?

Examples of data created by a business include:– Product data: inventory, description, pricing, availability, sales

numbers and projections

– Customer data: orders, shipping details

– Account data: banking, financial services industry

– Medical data: health care providers, insurance industry, hospitals


Data Creation: Business

Where is this data stored?

Business data can be stored on:– Employee workstations

– Servers

– Disk arrays

– Tapes

– CDROM/DVDs

– Off-site libraries


Value of Data to a Business: Information

What do businesses “do” with the data they collect?

They turn it into “information”.

Examples of information include:– Buying habits and patterns of customers

– GPS locations of delivery trucks

– Health history of patients

– Locations where a credit card is used


Value of Information to a Business

Identifying new business opportunities. For example:– Buying/spending patterns: Internet stores, retail stores, supermarkets.

– Customer satisfaction/service: tracking shipments & deliveries.

Identifying patterns that lead to changes in existing business. For example:– Reduced cost: delivery service optimizing utilization of vehicles and gas.

– New products: MP3 player speaker systems.

– New services: security alerts for “stolen” credit card purchases.

– Targeted marketing campaigns: communicate to bank customers with high checking account balances about a special savings plan.

Creating a competitive advantage!


Value of Individual Data to a Business

What data, created by individuals, might be valuable to a business?

Examples of business value from individuals’ data include:– On-line resume storage and management service .

– On-line photo storage and organizer.


Information Availability - Downtime is Expensive

Millions of US Dollars per Hour in Lost Revenue

6.5

3.6

2.8

2.6

2.0

1.6

1.6

1.5

1.3

1.2

1.1

Retail brokerage

Point of sale (POS)

Energy

Credit card sales authorization

Telecommunications

Call location

Manufacturing

Financial institutions

Information technology

Insurance

Retail

Source Meta Group, 2005


Types of Data

Structured

UnstructuredE-mail & Attachments

PDFs

Checks

X-rays

PaperDocumentsRich Media

Web PagesAudio & Video

RecordsInvoices

Manuals

Claims

Contracts Instant Messages

Forms

Images

XML

Rows and Columns

Data can be categorized as either structured or unstructured data.

Over 80% of enterprise information is unstructured. (Fulcrum Research)

What has been the traditional approach to storing all this data?


Storing Data: An Evolution

Centralized: terminals connected to a Mainframe computer which had connectivity to internal or external storage devices (disks, tapes).

Decentralized: With the advent of Open Systems, business units within an enterprise adopted a Client-Server model.

Centralized: Networked Storage (is the current “best practice” model being used in IT).


Worldwide Information Growth

~60% Average Growth Rate

>70% in 2005

0%

40%

60%

80%

100%

120%

2001 2002 2003 2004 2005e

Annual Growth of Data Stored on Disk Arrays

Data Source: IDC


Key Information Management Challenges

Planning for capacity growth

Classifying data

Address data availability

Security


Summary

Topics covered in this lesson included:

Who is creating data and the amount of data being created

The value of data to business

The challenges in data storage and data management

The solutions available for data storage


Data Storage Solutions Data Storage Solutions

Module 1.2


Data Storage Solutions


List the common storage media and solutions.

Describe the three common storage environments.


Lesson: Storage Solution Alternatives

Upon completion of this lesson, you will be able to:

Describe the different media and solutions available to address data storage needs of a business.

Describe the role of each solution in the overall data storage needs.

Describe the advantages of disk arrays.


Common Data Storage Media

Tape Library: A collection of tape drives and tapes

Jukeboxes: A collection of optical disks and drives

Disk Arrays: A collection hard disks

Each solution addresses specific needs for data storage and management.– Tape Library – Backup/Restore; Archival of data

– Jukeboxes – Typically to store non-changing content over long periods of time

– Disk Arrays – To store data that has to be immediately accessible and on-line


Tape Storage Systems

• Tape based storage is relatively inexpensive, compared to disk arrays.

• They served as primary storage solutions in the early days.

• Tape drives use Read/Write heads to record bits of data onto magnetic material on the tape surface.

• This technology continues to evolve, providing higher storage capacity, greater reliability, and improved performance.


Storing Data on Tape

Data is recorded sequentially from the beginning to end, one byte after another.

Because data is stored linearly along the length of the tape, random access to specific bits of data is slow and time consuming. This severely limits tape as a medium for real-time, rapid access to data.

Tapes cannot be shared among multiple users or applications.


Optical Data Storage

• Popular in small, single-user computing environments.

• Frequently used by individuals to store and share data, or as backup solution.

• Also used as distribution medium for applications, or as a means of transferring small amounts of data from one self-contained system to another.


Disk Based Storage

From a historical perspective, we will explore the following disk based storage solutions:

• DASD: Direct Access Storage Device

• JBOD: “Just a Bunch Of Disks”

• Disk Arrays

• “Intelligent” Disk Arrays


Types of Disk Systems: DASD

Mainframe

Disk

• Introduced by IBM in 1956.

• The ‘oldest’ of the techniques for accessing disks from a host computer.

• Disks are accessed directly by a single host, historically a mainframe system.


Types of Disk Systems: JBOD

Host

Disk Disk Disk Disk Disk

• Multiple physical disks in an external cabinet.

• Array connects to a single server only.

• Provides higher storage capacity with increased number of drives.

• Data not protected by JBOD

Array


Types of Disk Systems: Disk Arrays

Host B

Disk 1 Disk 2 Disk 3 Disk 4 Disk 5

Disk ArrayController

Array controllers for optimized I/O operations and RAID (Redundant Array of

Independent Disks) calculations.

Higher speed interconnects between drives than JBODs.

Multiple host I/O channels.

Can be partitioned to allow each host to access its own set of drives.

Host A Host C

Host A Host B Host C


Types of Disk Systems: “Intelligent” Disk Arrays

Disk 1 Disk 2 Disk 3 Disk 4 Disk 5

Disk ArrayController

Highly optimized for I/O processing.

Cache for improving I/O performance.

Operating environments provide: Intelligence for managing Cache,

Array resource allocation (Logical Unit),

Host access to Array resources,

Connectivity for heterogeneous Hosts

Host BHost A Host C


Lesson: Summary


Tape

Optical

Disks – DASD

– JBOD

– Disk Arrays

– Intelligent Disk Arrays


Lesson: Storage Environment

Upon completion of this lesson, you will be able to:

Describe Direct Attached Storage (DAS) features.

Describe Storage Area Network (SAN) features.

Describe Network Attached Storage (NAS) features.


Direct Attached Storage – DAS

Client 2

Server AApplication A

Server BApplication B

Server CApplication C

Disks for Server A

Disks for Server B

Disks for Server C

Client 3

Client 1

Local Area

Network

SCSI


Network Attached Storage – NAS

Disks for File System A

Disks for File System B

NAS Device AFile System A

NAS Device BFile System B

Internal/External connectivityto disks or arrays

Server AFile System A

Server BFile System B

Client 1

Client 2

Client 3

Local Area

Network

Linux

Windows


Storage Area Network – SAN

Client 2

Client 3

Client 1 Server AApplication A

Server BApplication B

Server CApplication C

Disks for Server A

Disks for Server B

Disks for Server C

SANLocal Area

Network

Disk Array

Fibre Channel

FC Switch

Data Block


Lesson: Summary


Direct Attached Storage (DAS) features.

Network Attached Storage (NAS) features.

Storage Area Network (SAN) features.

Differences: Connectivity


Module Summary

Key points covered in this module:

The three types of data storage media.

The three storage environments.


Check Your Knowledge

What are the three most common storage media?

What are the three types of disk-based solutions?

What are the advantages of a disk-based storage solution?

What are the three storage environments?


Data Center InfrastructureData Center Infrastructure

Module 1.3

Computing Center Data Center


Data Center Infrastructure


List the five core elements of a Data Center infrastructure.

Describe the role of each element in supporting business activities.

Describe the requirements of storage systems for optimally supporting business activities.

Describe the challenges and activities in managing the storage systems of a data center.


The Core Elements

Applications

Databases – Database Management System (DBMS) and the physical and logical storage of data

Servers/Operating Systems

Networks (between clients and servers or between servers and storage)

Storage Arrays


An Example: order entry system

Consider an order processing system consisting of:

Application for order entry.

Database Management System (DBMS) to store customer and product information.

Server/Operating System on which the Application and Database programs are run.

Networks that provide– Connectivity between Clients and the Application/Database Server– Connectivity between the Server and the Storage system.

Storage Array.

Local AreaNetwork

Storage Area Network

Storage Array

Client

Server

Application User

InterfaceDatabase


An Example ..Closer Look

A customer order is entered via the Application User Interface on a client.

Local AreaNetwork


Client

Server

Application User

Interface

Storage Array



A customer order is entered via the Application User Interface on a client

The client accesses the Server over a Local Area Network.


Client

ServerStorage Array

Local AreaNetwork



A DBMS uses the operating system on the server to read and write this data to the physical location on a disk.


Client

Server

O/S and DBMS

Storage Array

Local AreaNetwork



A DBMS uses the operating system on the server to read and write this data to the physical location on disk.

A Network provides the communication link between the server and the storage array, and transports the read/write commands and data between the server and the storage array.


Client

ServerStorage Array

Local AreaNetwork



A DBMS uses the operating system on the server to read and write this data to the physical location on disk.

A Network provides the communication link between the client and the server, and transports the read/write commands and data between the server and the storage array.

A storage array receives the read/write commands and data from the server and performs the necessary operations to store the data on the physical disks.


Client

ServerStorage Array

Local AreaNetwork

Database


An Example.. Optimal Order Processing

The Application should be optimized for fast interaction with the DBMS.

The tables in the Database should be constructed with care so that the number of read/write operations can be minimized.

The Server should have sufficient CPU and memory resources to satisfy Application and DBMS needs.

The Networks should provide fast communication between Client and Server, as well as Server and Storage Array.

The Storage Array should service the read/write requests from the Server for optimal performance.


An Example.. A Final Look at Data Access

When the DBMS receives a request from the Application:

It first searches the Server memory. If data is found there, the operation takes, perhaps, a millisecond.

If not, it then uses the Operating System to request the data from the Storage Array.

Dedicated high speed networks transport this request to the Storage Array.

Intelligent Storage Arrays can deliver the requested data within a few milliseconds. They are also typically configured to protect data in the event of drive failures.


Key Requirements of (Intelligent) Storage Systems

AvailabilityAvailability

Data IntegrityData Integrity SecuritySecurity

CapacityCapacity

ScalabilityScalability

PerformancePerformance

ManageabilityManageability

Applicable to all elements of the Data Center Infrastructure.They are qualities that must exist for successful use of data.

Downtime per year99.%=3.7days99.9%=9hours99.99%=53minutes99.999%=5minutes

Authorized users

Physically relocate or logically reassign resources to supporting different critical business needs. (DBMS)

Smoothly increase or decrease resources(apps, DB, server, storage) as needed (business grow).

Meeting user expectations fortimeliness and response toI/O requests.More server, one storage array

I/O chain checks

Flexible to configureand monitor thestorage system.


Some Constraints to Meeting the Requirements

Constraints include:

Cost : the budget

Physical Environment: the site

Maintenance and Support: human resource

Compliance – Regulatory & Legal: the business rule

Hardware and Software infrastructure

Interoperability and Compatibility


Management Activities

Data Center management activities include:

Provisioning / Capacity Planning / Resource Planning

Monitoring

Reporting


Monitoring

Performance

Security

Data Protection

Utilization

– Attempts and types of intrusions

– Data transfers or transactions per minute

– Hardware errors that are detected and corrected

– Numbers of users and resource use (CPU, Memory, Storage)

– Network traffic


Reporting

Utilization

Performance

Reports help in trend analysis. For example, periodic

reporting on disk space utilization enables predicting storage

capacity requirements in the future. Many organizations use

reporting as a chargeback system to recover cost of providing IT

infrastructure services to their internal customers. In turn, the

users may be held accountable for the resources they consume.


Provisioning: provide the hardware, software and resources needed to run the Data Center

Buy hardware

Buy software

Installing hardware and software

Training resources

Performance tuning the interrelated system


Capacity Planning

Understanding the business model – helps estimate the growth and support needs for the business. Understanding the business will help anticipate the data needs and growth in data capacity requirements.

Understanding the data life cycle for the business – helps identify the various stages of data in the business, requirements for migrating data to archived backup, and anticipate or plan for the increasing capacity needs.

Understanding changes in storage technology – ability to introduce new and more efficient storage methods to meet future capacity needs and manage costs.


Resource Planning

Understanding the procedures and tasks in the data center.

Changes in policy, procedures, or business needs.

Availability of qualified candidates.

Ability to train data center resources.

Sufficient budget to supply staff 24 X 7.


Module Summary

Key points covered in this module:

The five core elements of a Data Center Infrastructure.

The role of each element in supporting business activity was explained with an example of an order entry process.

The importance of an Intelligent Storage Array.

Key requirements of storage systems to support business activities as well as some of the constraints.

Management activities in a data center operation with focus on storage systems.


Check Your Knowledge

What are the five core technology elements of the Data Center Infrastructure?

What are the seven requirements of storage technology?

What are the common management activities in a Data Center?


Who is the industry leader?

Major companies (IBM, HP, Sun, Microsoft,

Hitachi, EMC, Network appliance, even

Cisco) are competing for storage market!