You are on page 1of 62

Topic 1 - Introduction to Computer Hardware Architecture

PC Architecture (TXW102)
Topic 1:
Introduction to Computer Hardware
Architecture

2006 Lenovo

PC Architecture (TXW102)

February 2007

Topic 1 - Introduction to Computer Hardware Architecture

Objectives:
Computer Hardware Architecture
Upon completion of this topic, you will be able to:
1. Identify the types of computers and their key differentiating features
2. Define various PC architecture terminology including computer layers,
controllers, and buses
3. Identify common industry standards and the objective of benchmarks
used with computer systems
4. Identify two features to enhance computer security

2006 Lenovo

PC Architecture (TXW102)

February 2007

Topic 1 - Introduction to Computer Hardware Architecture

Types and Features of Computers

Notebook
- Mobile
- Wireless
Designs
- Ultraportable (one spindle)
- Full function (two spindles)
- Desktop alternative
(three spindles)
- Tablet
Brands
- Lenovo 3000
- Lenovo ThinkPad

Desktop
- Non-mobile
- Wired connection
Designs
- Mini Desktop
- Ultra Small
- Small
- Desktop
- Tower
- PC Blades
Brands
- Lenovo 3000
- Lenovo ThinkCentre

Server
- High security
- Data processing and storage
Designs
- Tower
- Rack
- 1U rack
- Blades

Brands
- IBM System x
- IBM BladeCenter

2006 Lenovo

Types and Features of Computers


The three main types of computers (or PCs) are
Notebooks
Desktops
Servers

PC Architecture (TXW102)

February 2007

Topic 1 - Introduction to Computer Hardware Architecture

Notebooks
Notebooks are optimized for traveling and for mobile users who need easy access to data.
Notebooks can be categorized as ultraportable (around three pounds), full function (up to seven
pounds), or a desktop alternative (desktop features in a larger notebook design). Lenovo uses the
brand names 3000 or ThinkPad for its notebook systems.

Lenovo 3000 C200

Lenovo ThinkPad Z61t

The tablet PC is another kind of PC. Tablet PCs are based on a Microsoft specification for inkenabled touch screen computers (using Windows XP Tablet PC Edition). The tablet PC comes in
two form factors: slate (which has no keyboard attached because the tablet can be connected to a
docking station) and convertible (includes integrated keyboard). Lenovo markets the ThinkPad X60
Tablet, which is a convertible.

Lenovo ThinkPad X60 Tablet

PC Architecture (TXW102)

February 2007

Topic 1 - Introduction to Computer Hardware Architecture

Desktops
Desktops are for users who work in one place and need access to data on the desktop or through a
network. Lenovo uses the brand names 3000 or ThinkCentre for its desktop systems.
Common desktop mechanical designs include the following:
Mini Desktop (0 slot x 2 bay)
Ultra Small (1 slot x 2 bay)
Small (2 slot x 3 bay or 4 slot x 3 bay)
Desktop (4 slot x 4 bay or 3 slot x 3 bay)
Tower (4 slot x 5 bay)

Tower
Mini
Desktop
Small

Desktop

Common Desktop Mechanicals

Lenovo 3000 Tower


(4 slot x 4 bay)

Small
(2 slot x 3 bay)
Desktop
(3 slot x 3 bay)
Desktop
(4 slot x 4 bay)

Ultra Small mechanical in ThinkCentre M55


(1 slot x 2 bay)

PC Architecture (TXW102)

Desktop Mechanicals Introduced in 2005

February 2007

Topic 1 - Introduction to Computer Hardware Architecture

Higher end desktop systems are commonly referred to as workstations. Workstations are often used
for graphics and engineering users. They may run a UNIX-based operating system instead of a
Microsoft Windows-based operating system.
Below is a comparison of PC workstations and UNIX workstations.
PC Workstations

UNIX Workstations

Basics

Minimum configuration of latest Intel


processor; price starting around
$2,000

Minimum configuration of Sun


UltraSPARC or IBM PowerPC
processor; more memory; higher-end
graphics

Strengths

Low price; familiar Windows interface


and Windows applications run
natively; ease of manageability

Strong graphics; large number of


third-party applications available for
engineering and financial modeling;
processor power to spare

Weaknesses

Still lags behind UNIX counterparts in


3D graphics capabilities; large
applications may overwhelm system
performance; lack of third-party
graphics applications

Difficult to manage; price prohibitive


in some cases; Microsoft Office
applications don't run natively

Servers
Servers are computers that provide services to other computers, called clients. Servers are in secure
areas because so many users are dependent on their function. They include file servers, print
servers, terminal servers, Web servers, e-mail servers, database servers, and computation servers.
Server designs include
Tower, which rests on the floor
Rack-based, which must be installed in a rack
1U rack-based, which is a 1U-high server that must be installed in a rack. 1U is equal to 1.75.
Server blades, which have server circuitry on a single board which slides into an enclosure with
other blades.
Note: Servers for racks vary in height by a U measurement (a U is 1.75-inch height). 1U servers
are popular for Web sites because for Web pages it is better to spread the load across multiple
servers (horizontal scalability) rather than to increase the processing power of a centralized server
(vertical scalability).

PC Architecture (TXW102)

February 2007

Topic 1 - Introduction to Computer Hardware Architecture

Types and Features of Computers:


Differentiating Computer Features

Notebook
Size and weight
Power mgmt/battery
Screen type and size
Integrated wireless
Docking station or port replicator
Number of spindles or bays
Modular bay(s)
Sales presentation capability
Integrated wireless, Bluetooth,
infrared
- Security chip
- ThinkVantage Technologies
-

Desktop
Fastest processor
Graphics performance
3D graphics adapters
Systems management
Removable storage (DVDRW)
Chipset
Security chip
ThinkVantage Technologies

Server
- Support many concurrent users

(up to 1000s)
- Multiple processors
- Large memory capacity
- Large disk capacity

(internal and external)


- Redundant components (disk,

fans, power supplies, memory,


etc.)
- Hot-swap components to
maximize uptime (disk, fans,
power supplies, memory, etc.)
- Predictive Failure Analysis to
warn the admin of any
impending failure
2006 Lenovo

Differentiating Computer Features


Each type of computer has important characteristics that distinguish it from each other.
Key differentiating features of notebooks are size and weight, power management and battery,
screen type and size, integrated wireless, docking station or port replicator, number of spindles or
bays, modular bay(s), sales presentation capability, integrated wireless, Bluetooth, infrared, security
chip, and ThinkVantage Technologies.
Key differentiating features of desktops are fastest processor, graphics performance, 3D graphics
adapters, systems management, removable storage (DVDRW), chipset, security chip, and
ThinkVantage Technologies.
Key differentiating features of servers are support of many concurrent users (up to 1000s), multiple
processors, large memory capacity, large disk capacity (internal and external), redundant
components (disk, fans, power supplies, memory, etc.), hot-swap components to maximize uptime
(disk, fans, power supplies, memory, etc.), and Predictive Failure Analysis to warn the admin of
any impending failure.

PC Architecture (TXW102)

February 2007

Topic 1 - Introduction to Computer Hardware Architecture

Types and Features of Computers:


PC Blades
User Ports
The ClearCube User Port connects
computer peripherals like the
monitor, keyboard, mouse, speakers,
or USB peripherals to PC Blades at
the data center or telecom closet.

PC is removed
from users desk
and replaced with
a small User Port

User Port (I/Port)

PC Blade is in a
rack in secure,
centralized location

Lenovo resells
ClearCube-branded
PC Blades and
management
software

Ethernet

User Port
(C/Port)

Direct Connection

PC Blade
The PC Blade is each user's actual
computer: a configurable, Intel-based PC
Blade that delivers full functionality to the
desktop.

Cage
The ClearCube Cage is
a centralized chassis that
houses up to eight PC Blades.

ClearCube Management Suite


The ClearCube Management Suite empowers
administrators to manage the complete ClearCube
infrastructure from any location. This powerful,
remotely accessible suite includes a versatile set of
features such as "hot spare" switching, move
management, and automatic data backup.

2006 Lenovo

PC Blades
PC Blades separate the guts of the PC from the physical desktop, putting processing power in data
centers and computer rooms. Employees then have only a monitor, keyboard and mouse on their
desks, along with a client appliance that is linked back to a blade server. PC blades offer a range of
benefits, including streamlined management and tighter security since all the hardware is
centralized. PC Blade configurations provide a dedicated blade to each user or a pool of blades that
can be dynamically allocated. In addition, spare blades can be used to provide hot backup to avoid
system outages.
PC Blade Advantages
Centralized asset management PC Blade hardware is centralized for easy access and asset
management.
Mission critical applications Blade infrastructure has high levels of redundancy; users can be
swapped to a functioning blade very quickly in case of hardware or software failure
Reduced support costs Hardware or software upgrades can be managed centrally in a fraction
of the time it would take to upgrade large numbers of dispersed PCs.
Multiple locations There is potential to support multiple locations with PC Blades by remotely
switching a user to a spare standby blade in the event of hardware failure (saving the cost of an
urgent engineer visit or keeping support staff on-site).
Reduced costs for new users It is lower cost to install and configure a new user with a PC
Blade than a desktop.

PC Architecture (TXW102)

February 2007

Topic 1 - Introduction to Computer Hardware Architecture

Easy relocation There are no significant costs when users move work location within a building.
Improved security The physical asset and intellectual property on the disk are centralized, e.g., it
is easier to steal a hard disk from a desktop than a PC Blade.
Reduced user down time Spare PC Blades can be configured to provide hot backup in case of
hardware failure.
Improved appearance In front office environments, the clients user port has no moving parts,
generates no noise, produces little heat, and requires less space.
Remote access Users can access their own PC environment from multiple desks in the building
or from other remote locations with blade infrastructure installed
PC Blade Disadvantages
Higher acquisition cost Purchase price of PC Blade and its infrastructure is higher than a stand
alone PC.
No wireless mobility Mobile users or users who need to work away from their desks are not
supported.
Lagging technology PC Blade processors and technology may be six to 12 months behind
desktop technology.
Unsuitable for power graphics users.
New infrastructure Significant change to current PC deployment, maintenance, and support
(skills, tools and processes).
More difficult to plan and manage upgrades when customer has a mix of PC Blades and traditional
desktops.
Lack of local CD and DVD drives except USB devices which open security risks and asset control
issues.
User resistance for advanced/experienced PC users to losing access to 'their' PC.
Extra cost for redundancy Extra cost for closet spare (with cooling) to enable redundancy.
Technology lock in Little option to cascade or sell blades to other users or customers.

PC Architecture (TXW102)

February 2007

Topic 1 - Introduction to Computer Hardware Architecture

Types and Features of Computers:


ClearCube PC Blade Products
PC Blade
- PC Blade is located with other
PC Blades in a rack in centralized
location
- Intel processor, memory, disk, and
graphics on PC Blade
PC Blade (ClearCube R1300)

User Port
- Small client device that connects the
users monitor, keyboard, mouse,
speakers, and USB devices to their
PC Blade
- No moving parts, generates no noise,
and creates little heat
- Can support multiple monitors

User Port (ClearCube C/Port)

2006 Lenovo

ClearCube
ClearCube is a company that has offered PC Blades since 1997 and dominates the PC Blade
market. Lenovo resells ClearCube-branded PC Blades and management software.
See www.clearcube.com for more information.

PC Architecture (TXW102)

February 2007

10

Topic 1 - Introduction to Computer Hardware Architecture

ClearCube Products
A PC Blade solution from ClearCube require PC Blades, User Ports, Cage, BackPack, management
software, and optional accessories.
The ClearCube PC Blade is an Intel-based computer that delivers full PC functionality to the user
from a centralized location. The PC Blade is very compact and house all the standard PC components
including an Intel processor, memory, disk drive, and graphics.

ClearCube R1300 PC Blade

For the client desk, each user requires a ClearCube User Port which is a compact device that
connects the users monitor, keyboard, mouse, speakers, and USB devices to their PC Blade. The
User Port has no moving parts, generates no noise, and creates very little heat. ClearCube offers
several User Ports as follows:
C/Port The C/Port connects to a centralized PC Blade over standard Category 5 cabling up to 200
meters long. The C/Port is about the same size and weight as a VHS cassette tape.
Fiber C/Port The Fiber C/Port connects to a centrally located PC Blade over a point-to-point fiber
optic cable up to 500 meters long. Because stray electromagnetic emissions are eliminated, the
Fiber C/Port is ideally suited for ultra-secure environments.
Multi-Video Expander For demanding, multi-task environments, the Multi-Video Expander is a
C/Port accessory that can be matched with an NVIDIA mezzanine video card on the PC Blade to
drive up to four displays. By accommodating analog and digital input monitors in a variety of
physical configurations, the Multi-Video Expander is the perfect choice for hectic, cramped trading
floors.
I/Port The I/Port lets users connect to their PC Blades over a standard switched ethernet network.
By connecting over ethernet and leveraging thin client technology, the distance between a PC
Blade and the desktop can be unlimited.

PC Architecture (TXW102)

February 2007

11

Topic 1 - Introduction to Computer Hardware Architecture

ClearCube C/Port (back)

ClearCube C/Port (front)

A ClearCube Cage is a 3U-high chassis which houses up to eight PC Blades. Each Cage is a selfcontained unit that provides all of the ethernet connections, airflow management, and power required
by the PC Blades. Up to 14 Cages (112 PC Blades) can be mounted in a standard 6 foot rack. A
BackPack connects directly to the back of each Cage, and PC Blades slide in the front.

ClearCube Cage

ClearCube Cages in a rack

The ClearCube Blade Switching BackPack is an interface that plugs into the rear of the Cage and
provides all of the external connections for the PC Blades. The BackPack contains the distance
extension circuitry that links each PC Blade to a User Port. The Blade Switching BackPack also
features a full 8x8 switching matrix and spare switching circuitry that controls the connections to
standard and Fiber C/Ports.

ClearCube Blade Switching BackPack

PC Architecture (TXW102)

February 2007

12

Topic 1 - Introduction to Computer Hardware Architecture

The ClearCube Management Suite (CMS) is a comprehensive set of software tools that work
together to give IT administrators unprecedented control over their computing assets. Inventory
management, automated health alerting, remote control, hard drive image management, data backup
and Blade switching are just some of the capabilities ClearCube offers. All of the CMS applications
can be accessed via secure web browser and can be integrated with industry-standard systems
management tools.

Switch Manager

Blade Manager

Data Failover

Move Manager

ClearCube Management Suite


User Port

Cage

Layer 3: System management software


BackPack

Layer 1: Industry standard components

PC Blade

PC Architecture (TXW102)

Layer 2: ClearCube infrastructure

February 2007

13

Topic 1 - Introduction to Computer Hardware Architecture

Types and Features of Computers:


Lenovo 3000 Desktops and Notebooks
Lenovo 3000 Desktops (J Series)

Small Desktop

Tower

Lenovo 3000 Notebooks

C200 15" Regular

N100 14.1" Wide and 15.4" Wide

V100 12" Wide


2006 Lenovo

Lenovo 3000 Desktops and Notebooks


In February 2006, Lenovo introduced the 3000 Family of desktops and notebooks. The
following information shows the branding and positioning of the two product lines.
Think Family
(ThinkPad and ThinkCentre)
The Ultimate Business Tool

Lenovo 3000 Family


The Smart Choice for
Small Business

Common Attributes

Success, Prestige
Rock Solid
Lowest Total Cost

of Ownership

Innovative
High Quality and Reliability
Service Excellence

Large Enterprise/Mid-Market

Worry-free
Great Value
Stylish Design

Small Business/Very Small Business

The Lenovo 3000 J100/J110 features Intel processors while the Lenovo 3000 J105/J115
features AMD processors.
The Lenovo 3000 notebooks are available in various screen sizes in both widescreen and
standard screen formats.
Visit lenovo.com/3000 for more information on the Lenovo 3000 Family of desktops and
notebooks.

PC Architecture (TXW102)

February 2007

14

Topic 1 - Introduction to Computer Hardware Architecture

Types and Features of Computers:


Lenovo ThinkPad, ThinkCentre, and ThinkVision
Lenovo Think Family

ThinkVantage
Technologies

ThinkPlus
Accessories
and Services

ThinkVantage
Design

ThinkPad, ThinkCentre, and ThinkVision offerings will continue to differentiate


Lenovo from our competitors with
Quality, service, and support expected from Lenovo
Industrial design that simplifies and enhances usability
Open standards-based products that work well together
Lenovo innovation that delivers key benefits for customers
2006 Lenovo

Lenovo ThinkPad, ThinkCentre, and ThinkVision


The Lenovo Think-branded family of offerings includes the following brands:
ThinkPad Notebook category
ThinkCentre Desktop category
ThinkVision Visuals category
ThinkVantage Technologies Solutions and offerings category
ThinkPlus Accessories Accessories and upgrades for Think products
ThinkPlus Services Services for Think offerings
An essential part of what makes a product a ThinkPad, ThinkCentre, or ThinkVision offering is
its industrial and graphic design. Lenovo calls this design approach ThinkVantage Design.
ThinkVantage Design is built upon the concept of synergistically joining form and function.
ThinkVantage Design provides value to the customer by providing meaningful innovation that
enhances the ownership experience. It also has its own design DNA based on the classic
ThinkPad design, which is its heritage.
Visit lenovo.com for more information on Lenovo brand offerings.

PC Architecture (TXW102)

February 2007

15

Topic 1 - Introduction to Computer Hardware Architecture

Types and Features of Computers:


ThinkVantage Technologies
ImageUltra Builder

System Migration Assistant

Consolidates multiple software


images into one master image

Moves system settings and data


easily from an old PC to a new PC

Software Delivery Center

Access Connections

System Information Center

Client Security Solution

Automates delivery of application


software updates to PCs

Switches painlessly between


settings for different wireless
and wired networks

Collects and tracks PC asset and


security compliance information

Secures users PCs, data, and


network communications from
unauthorized access

System Update

Rescue and Recovery

Accesses, downloads, and installs the


latest updates for Think systems

Enables hassle-free recovery


of data and system image

Productivity Center

Provides users with access to


self-help support tools and
information with just one click
2006 Lenovo

ThinkVantage Technologies
ThinkVantage Technologies are a select group of offerings from
Lenovo designed to address emerging customer needs. Adding
value to open industry standards, ThinkVantage Technologies help
customers manage the cost of deploying end-user systems,
implement new technologies such as wireless computing, and help
ensure that these technologies can be implemented securely. While
many of these offerings currently exist, some are being significantly
enhanced and all of them have now been consolidated into a single
family of offerings.
Visit lenovo.com/thinkvantage for more information.

Popup from ThinkVantage


Productivity Center

PC Architecture (TXW102)

February 2007

16

Topic 1 - Introduction to Computer Hardware Architecture

Types and Features of Computers:


Benefits of ThinkVantage Technologies
ThinkVantage Technologies address the entire customer ownership experience from
deployment to disposal.

Image Creation
ImageUltra Builder
- Hardware Independent
Imaging Technology
- Dynamic Operating
Environment
- Software Delivery
Assistant

Hassle Free
Connection

Secure Client
Data

Access
Connections

Client Security Solution

Productivity Center

Hard Drive
Data
Destruction

Password Manager

Rescue and
Recovery

Secure Data
Disposal

Secure Data
Media

Information
and Asset
Management

Image on Demand

EndEnd-user SelfSelfhelp Portal

Active Protection
System

Imaging Technology
Center

System Information
Center

Backup and
Recovery

Network Deployment
Remote Deployment
Manager

Rescue and Recovery

Client Migration

Critical Updates

System Migration
Assistant

Rescue and Recovery


with Antidote Delivery
Manager

Software on
Demand
Software Delivery
Center
System Update

2006 Lenovo

Benefits of ThinkVantage Technologies


Industry analysts state that the annualized cost of a PC represents less than 20 percent of the annual
total cost of ownership. ThinkVantage Technologies address the other 80% to help reduce your
total cost of ownership. ThinkVantage Technologies also help improve your business' productivity
and efficiency throughout each system's life cycle as you deploy, connect, protect, support, and
dispose of your company's PCs.

End-user productivity
(value out of the box that also can
provide key IT benefits)
Access Connections
Productivity Center
Active Protection System
Client Security Software
Rescue and Recovery
System Migration Assistant
System utilities

PC Architecture (TXW102)

Life-cycle Management
(solutions for SMB and LE)
System Information Center
Software Distribution Center
ImageUltra Builder
Secure Data Disposal
Remote Deployment Manager

February 2007

17

Topic 1 - Introduction to Computer Hardware Architecture

Life Cycle Phase

TVT

Purpose/Function of Tool

Deploy

ImageUltra Builder

Umbrella name for imaging technology that is focused on simplifying the


complexity of creating and managing corporate images; ImageUltra
Builder consists of the three components described below

- Software Delivery Assistant


(SDA)

Provides customized installation of software applications based on a


user's unique work group assignment and/or needs

- Dynamic Operating Environment


(DOE)

Consolidates support for multiple operating systems and languages into


one Super Image

- Hardware Independent Imaging


Technology (HIIT)

Provides hardware-independent images that will support multiple system


types via system-specific drivers and applications pulled from the PCs
service partition (supported on ThinkPad and ThinkCentre systems only)

System Migration Assistant (SMA)

An easy-to-use migration tool that automates the migration of both


settings and data through a menu system and advanced scripting
capability

Remote Deployment Manager


(RDM)*

Network-based imaging and system support tool that distributes images


and updates system settings (includes PowerQuest Drive Image Pro Lite)

Connect

Access Connections

Manages all wireless and wired connectivity settings and allows easy
switching between them

Protect

Client Security Solution (CSS)

Security software that provides authentication of end-user identity,


encryption of data, and simplified password management

Active Protection System (APS)

Prevents some hard-drive crashes on most new ThinkPad models by


temporarily stopping the hard drive when a fall or similar event is
detected; provides up to four times greater impact protection than
systems without this feature

Productivity Center

Provides one-button access to self-help support tools and information


about a users system

Rescue and Recovery

A help desk behind the button that allows a system to recover itself from
OS corruption and even hard drive failures, fills the gap between
traditional backup and restore programs and re-imaging, allows remote
system recovery with or without user intervention, and automates the
deployment of critical patches, even if a system will not boot.

System Information Center (SIC)

An electronic inventory management solution that tracks client PC


hardware and software assets, provides ThinkVantage Technology usage
information, and reports and measures security compliance

Software Delivery Center (SDC)

Automates delivery of application software updates to PCs without enduser intervention or disruption

Asset ID

A unique asset-tagging technology that allows data, events, actions, and


responses to be read by or become interactive with other programs

System Update

Accesses, downloads, and installs the latest updates for Think systems

IBM Director Agent

An advanced CIM/WMI-based agent for systems and asset management


that allows extensive upward integration into other enterprise
management tools and databases

IBM Director

A powerful systems management program with extensions targeted for


improved system setup, remote manageability, and event monitoring of
clients and servers

Secure Data Disposal

An automated program that allows multiple levels of disk cleansing, which


ensures systems are properly safeguarded during disposal; meets the
DOD Level 5 and German standards for safe disposal

Support

Dispose

PC Architecture (TXW102)

February 2007

18

Topic 1 - Introduction to Computer Hardware Architecture

ThinkVantage Technology

Potential Savings*

Assumptions Used in Calculating Savings

ImageUltra Builder

$100 per unit

Deployed once per system versus typical cloned


image management and loading process

System Migration Assistant

$70 per unit deployed

Deployed once per system versus manual


processes

Remote Deployment Manager

$90 per system

Deployed once per system versus manual


processes

Access Connections

$50 per wireless PC

Annual savings and only in notebook systems


Assumes two help desk call per user per year

Client Security Solution

$124-250+ per unit

$35-60 hardware replacement


$49-150 encryption software replacement
$40 support cost reduction
Replaces comparable equipment (key fobs, etc.)

Active Protection System

Rescue and Recovery

$200 or more per


occurrence

Only in notebook systems

$180 per occurrence

Used for one incident in 13% of installed systems

$200 is the hardware replacement cost for a


ThinkPad 30 GB hard drive

Average support time savings of 183 minutes


Secure Data Disposal

$45 per PC

Once per PC per life cycle

* Potential savings are based on typical customer environments. Some figures represent costs that customers may redirect from
labor-intensive areas to other areas of their business. Other figures are based on cost avoidance of competitive solutions
purchased separately. All figures are calculated using the TVT and Wireless Calculators and data from Gartner Research and
customers. Actual savings are not guaranteed and will vary by customer.

PC Architecture (TXW102)

February 2007

19

Topic 1 - Introduction to Computer Hardware Architecture

PC Architecture:
Computer Layers
Layered structure

User

- This structure allows for


compatibility.

Applications

- Bypassing layers increases


performance.

BIOS (basic input/output system)


- Located in flash memory
(sometimes called EEPROM)
- Supports plug-and-play

Middleware API
Operating
system
Device
driver
Firmware

BIOS

Adapter

Hardware

Layers

- Supports power management

Device driver
- Software to control a piece of
hardware

EEPROM BIOS
2006 Lenovo

Computer Layers
A computer consists of several layers that each have interfaces to communicate to the layer next
to it. A layered structure allows for compatibility; for example, the same shrink wrapped
operating system can work on millions of PCs from different vendors because it interfaces with
industry standard BIOS calls. The disadvantage to the layers is that each layer can slow
performance. So to increase performance a layer could be bypassed; for example, an application
could be written directly to the BIOS and device driver of a specific computer which would gain
performance, but would only work on that unique computer.
Some of the different computer layers shown in the diagram above are explained below.
Applications are the software programs with which a user typically interacts, such as those used
for word processing (Microsoft Word), Web browsing (Internet Explorer), sending e-mail
(Lotus Notes), and using spreadsheets (Microsoft Excel).
Middleware is software that provides an additional level of abstraction to applications. The idea
behind the middleware is to hide the complexity of implementing code that is not strictly
related to the business objectives that the application is supposed to be written for. Writing
applications against the basic APIs that the OS is able to expose is sometimes very time
consuming and it might take a while before a programmer starts to get into the business
modules of the application being developed. Using middleware is like actually talking to a
cleverer interface compared to the interface provided by the OS. Middleware has to implement
all the boring stuff, so that developers can concentrate on the business logic. Examples
include IBM DB2, Oracle, Microsoft SQL Server, Lotus Domino, Microsoft Internet
Information Server, IBM WebSphere, and BEA WebLogic.

PC Architecture (TXW102)

February 2007

20

Topic 1 - Introduction to Computer Hardware Architecture

The operating system is a set of programs that provides an environment in which applications
can run, allowing them easily to take advantage of the processor and I/O devices, such as disks
or adapters. Examples include Windows 2000, Windows XP, Vista, Red Hat Linux, and AIX
5L.
The Basic Input/Output System (BIOS) is a set of program instructions that activates system
functions independently of hardware design (the layer between the physical hardware and the
operating system) and allows for software compatibility. BIOS is typically located in flash
memory (EEPROM) on the systemboard. When a PC is started, the BIOS runs a power-on selftest (POST). It then tests the system and prepares the computer for operation by searching for
other BIOSes on the plug-in boards and setting up pointers (interrupt vectors) in memory to
access those routines. It then loads the operating system and passes control to it. The BIOS
accepts requests from the drivers as well as the application programs. The BIOS supports plugand-play and power management. BIOS vendors include IBM, AMI, and Phoenix. Although
there are several BIOS vendors, there are few differences among their products.
Note: To preclude the problem of performing OS, BIOS, or driver updates before the OS or
network drivers are loaded, a Preboot Execution Environment (PXE) allows the system to boot
off the network. At boot, a PXE agent executes, and the PC gets an IP address from a DHCP
server and then uses the BOOTP protocol to look for a PXE server. The PXE client is firmware
implemented in BIOS (if LAN hardware is on the systemboard) or as a boot PROM (if LAN
adapter). Programs, including those in the PXE environment, require system configuration and
diagnostic information. A Systems Management BIOS (SMBIOS) is a chip that makes the
necessary information available via BIOS calls that are available through the OS and in the
preboot environment.
Firmware is usually the layer of software that is between the device driver and adapter. It
typically is on a EEPROM of an adapter card and can be upgraded with new releases. Firmware
is similar to BIOS.
Device Drivers are a type of software (which may be embedded in firmware) that controls or
emulates devices attached to the computer such as a printer, scanner, hard disk, monitor, or
mouse. Device drivers are typically loaded low into the memory of PCs at boot time. A device
driver expands an operating system's ability to work with peripherals and controls the software
routines that make peripherals work (a network card, a disk, printer). These routines may be part
of another program (many applications include device drivers for printers), or they may be
separate programs. Basic drivers come with the operating system, and drivers are normally
installed for each peripheral added.

PC Architecture (TXW102)

February 2007

21

Topic 1 - Introduction to Computer Hardware Architecture

Layers of a PC: Illustration


To illustrate the different layers of a PC, below is a graphic that shows the series of processes
that occur when a user executes a single keystroke on a computer keyboard, and the
corresponding key to the numbered steps.

Application Program

6.
Operating System

4.
7.
Keyboard
Cable

1.

5.

BIOS 3.

Video
Circuitry
Keyboard Port
Hardware

2.

Hardware

1. The user presses N on keyboard.


2. The code corresponding to pressed key is sent over the cable to the keyboard port on
the systemboard.
3. The keyboard BIOS routine accepts the code, translates it into the letter n, and passes
it to the operating system.
4. The operating system passes the keystroke to the application program and sends the n
to the video BIOS (or the device driver).
5. The video BIOS sends the n to the graphics circuitry.
6. The application program accepts the keystroke and instructs the operating system to
look for the next keystroke.
7. The n appears on the screen.

PC Architecture (TXW102)

February 2007

22

Topic 1 - Introduction to Computer Hardware Architecture

PC Architecture:
Subsystems
Major internal subsystems of a PC:

Processor (Core 2 Duo)


L2 cache (2 MB)
Memory (2 GB)
Bus(es) (PCI, PCIe)
Graphics controller (SVGA)
Disk controller (SATA)
and disk (250 GB)

Slots (PCI Express)

PCI Express
x16 slot
PCI Express slots
Direct
Media
Interface
PCIe

Processor +
L1/L2 cache
MCH or
GMCH
host bridge

Low Pin
Count interface

Memory

I/O
Controller
Hub
(ICH)

PCIe controller
PCI controller
SATA controller
IDE controller
USB controller

Firmware
hub

USB 2.0
AC '97 codec
or
High Definition Audio

4 SATA disks
Super I/O

Memory and optional


graphics controller

2006 Lenovo

Subsystems
Subsystems in a PC communicate to each other via buses. Buses adhere to a particular
architecture (set of rules) to allow compatibility with the numerous subsystems that adhere to
the same architecture.
Most PCs are associated with the term Wintel, which refers to Microsoft Windows and Intel
chip technologies. PCIe stands for PCI Express.
The processor is the central component of a PC. Intel and AMD are the main processor vendors
used in PCs.
Data in the processor, caches, memory, buses, disk controller, and graphics controller is stored
electrically; so when electrical power is shut down, this data is lost. Data on the disk is stored
magnetically, so the data is saved even when electrical power is removed.

PC Architecture (TXW102)

February 2007

23

Topic 1 - Introduction to Computer Hardware Architecture

PC Architecture:
Controllers
All major subsystems have controllers.
Controllers are circuitry controlling manner, method, and speed of access
to device.

Controllers are part of chipset.

Controller

Controls

Examples

L2 cache controller

L2 cache

2 MB

Memory controller

Memory

2 GB

Bus controller(s)

Data bus

PCI Express (PCIe)

Graphics controller

Monitor

ATI Radeon 9600

Disk controller

Disk

250 GB Serial ATA disk

2006 Lenovo

Controllers
All major subsystems have controllers that define how data will be obtained and stored.
Sometimes a controller is a single chip with the data stored in separate physical circuitry. For
example, a memory controller controls memory, but the data is stored in different physical chips
called DIMMs.
Sometimes a single physical chip contains multiple controllers. For example, the I/O Controller
Hub (ICH) is a single physical chip which houses the PCI Express controller, PCI controller,
Serial ATA controller, EIDE controller, USB controller, and other controllers.
Controllers are normally included in the chipset of the computer.

PC Architecture (TXW102)

February 2007

24

Topic 1 - Introduction to Computer Hardware Architecture

PC Architecture:
Buses
Most transfers use three buses

Control bus
Address bus
Data bus

16-bit bus = 16 wires for on/off charges (data)


32-bit bus = 32 wires for on/off charges (data)
64-bit bus = 64 wires for on/off charges (data)

I/O
Controller

Control

Data

Address

Processor

Data

Memory
Disk

Graphics

LAN

Some architectures multiplex signal on same bus (wires)


2006 Lenovo

Buses
If two subsystems are on a bus, such as in the diagram with processor and memory, a data
transfer first involves sending the address on the address bus. Next, data is sent on the data bus.
If multiple subsystems exist on a bus, a control bus is needed in addition to the address and data
bus. The control bus is used to signal which subsystem will control the bus for the next transfer.

PC Architecture (TXW102)

February 2007

25

Topic 1 - Introduction to Computer Hardware Architecture

Address Bus
An address bus determines how much memory the processor or any subsystem can directly
address. For example, a 32-bit address bus means 2 to the power of 32 or 4 billion unique
numbers to address 4 GB of memory.

0
0
0
0
0

0
0
0
0
0

0
0
0
0
0

0
0
0
0
0

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

0
0
0
0
0

0
0
0
0
0

0
0
0
1
1

0
0
1
0
0

0
1
0
0
1

Memory Addressing Similar to Car Odometer

Before data is read or written by a processor, the address of that data is sent first. This address is
sent on a separate set of physical wires called the address bus. The data is then sent on a
different set of physical wires called the data bus.
A processor is designed to use a certain maximum quantity of address lines. The amount of
physical memory that a processor can address is determined by this quantity. The number of
unique numbers that can be made by a base two number system (0s and 1s) with the quantity of
address digits determines the maximum addressable memory of a processor. Software can limit
this maximum addressability, for example, DOS sets the processor to use 20 address lines as
DOS only addresses 1 MB of memory.
Following are some processors and their addressability:
Address lines

Addressable memory

Examples

24

16 MB

486SLC

32
36
40
44

4 GB
64 GB
1 TB
18 TB

486DX2
Pentium 4, Xeon
EM64T physical memory
Itanium

48

256 TB

EM64T virtual memory

64

18 EX

IA-64 64-bit flat addressing

Sometimes operating systems limit addressability, so that the operating system can not utilize
all the available physical memory.

PC Architecture (TXW102)

February 2007

26

Topic 1 - Introduction to Computer Hardware Architecture

PC Architecture:
Bus Speeds
Processor +
L1/L2 Cache
System bus
400 to 1066 MHz
PCI Express
x16 slot
PCI Express slots

PCIe 2.5 GHz

MCH or
GMCH
Host Bridge

Memory
Memory bus
200 to 400 MHz

PCIe 2.5 GHz


PCI 2.0 33 MHz
PCI 2.0 slots

4 SATA
disks
Super I/O

I/O
Controller
Hub
(ICH)
Firmware
Hub

Direct Media Interface (DMI)


100 MHz

USB

Low Pin Count (LPC) Interface 33 MHz

Each bus is clocked at a different rate.


Bus speed is different from data transfer rate (MB/s).
Newer buses are double data rate (same MHz doubles throughput).
System bus and memory bus can be asynchronous or synchronous.
2006 Lenovo

Bus Speeds
Each bus in a PC has a speed (measured in megahertz) and a data transfer rate.
The bus between the processor and the memory controller was originally called the frontside
bus; the processor had a separate bus to its integrated L2 cache called the backside bus and a
separate bus outside the processor to the memory controller called the frontside bus. The
frontside bus and the backside bus were two different buses. With the introduction of the
Pentium 4 and follow-on processors, the frontside bus was named system bus, although both
terms were still used interchangeably. The change of the name to system bus was due to the fact
that the L2 cache was not isolated off a separate, independent bus to the degree that it was for
earlier processors, such as the Pentium II and Pentium III.
The memory bus is clocked at 200 to 400 MHz, but most memory today is double data rate
(DDR); this means data is transferred on both the rising and falling edge which doubles the
throughput from the base clock speed.
The system bus and the memory bus can be either synchronous or asynchronous, depending on
the memory controller of the chipset. Some memory controllers only support synchronous
system and memory bus speeds; some support either synchronous or asynchronous speeds. An
example of a memory controller that supports synchronous system and memory bus speeds is a
400 MHz system bus with a 200 MHz memory bus with PC2-3200 DDR2 memory (there is an
even multiple of 200 among 200 MHz and 400 MHz). An asynchronous example is a 400 MHz
system bus with a 266 MHz memory bus with PC2-4200 533MHz DDR2 memory.

PC Architecture (TXW102)

February 2007

27

Topic 1 - Introduction to Computer Hardware Architecture

In 1996 and 1997, the PC industry standardized the 66 MHz system bus. Migration to a 100
MHz system bus occurred in 1998, then to a 133 MHz bus in 2000. The Pentium 4 introduced a
400 MHz system bus in late 2000, although it was really 100 MHz 4 to yield 400 MHz. Later,
Pentium 4 processors utilized an 800 MHz system bus (200 MHz 4) followed by a 1066 MHz
system bus (266 MHz 4).
Data Transfer Rates
Data transfer rates (assuming that data is transferred on only one edge of the clock):
8-bit at 133 MHz (double data rate) in 266 MB/s (Hub Interface)
32-bit at 33 MHz is 132 MB/s (PCI bus)
32-bit at 66 MHz is 264 MB/s (PCI bus)
64-bit at 33 MHz is 264 MB/s (PCI bus)
64-bit at 66 MHz is 528 MB/s (PCI bus and system bus)
64-bit at 100 MHz is 800 MB/s (system bus)
64-bit at 200 MHz is 1.6 GB/s (backside bus to L2 cache; PC1600 DDR memory)
64-bit at 266 MHz is 2.1 GB/s (PC2100 DDR memory)
64-bit at 400 MHz is 3.2 GB/s (backside bus to full speed L2 cache, Pentium 4 system bus)

PC Architecture (TXW102)

February 2007

28

Topic 1 - Introduction to Computer Hardware Architecture

PC Architecture:
Cache
Cache is a buffer between subsystems.
Disk transfer could involve five cache locations.

PCI Express
x16 slot
PCI Express slots

PCI
Express
4.
SCSI

Processor +
1. L1 Cache
2. L2 Cache
MCH or
GMCH
Host Bridge

I/O
Controller
Hub
(ICH)

3. Memory

EIDE Disks

5.
Firmware
Hub

2006 Lenovo

Cache
Cache is a storage place (buffer or bucket) that exists between two subsystems in order for data
to be accessed more quickly to increase performance. Performance is increased because the
cache subsystem usually has faster access technology and does not have to cross an additional
bus. Cache is typically used for reads, but it is increasingly being used for writes as well.
For example, getting information to the processor from the disk involves up to five cache
locations:
1. L1 cache in the processor (memory cache)
2. L2 cache (memory cache)
3. Software disk cache (in main memory)
4. Hardware disk cache (some disks may only use an FIFO buffer)
5. Disk buffer
For reads, one subsystem will usually request more data than what is immediately needed, and
that excess data is stored in the cache(s). During the next read, the cache(s) is searched for the
requested data, and if it is found, a read to the subsystem beyond the cache is not necessary.

PC Architecture (TXW102)

February 2007

29

Topic 1 - Introduction to Computer Hardware Architecture

Industry Standards:
Restriction of Hazardous Substances (RoHS) Directive
Restriction of hazardous substances in electrical
and electronic equipment

Started by the European Union


Effective for shipments after July 1, 2006
Lenovo PC products comply across all of its
product lines worldwide

Restricted substances:

Lead
Mercury
Cadmium
Hexavalent chromium
Polybrominated biphenyls
Polybrominated
diphenylethers

FIN
FIN

IS

NO NO

RUS
ES

SE

LV
LT

IE

PT

PT SP

IE

SP

ES
LV
LT
BY

BY

GB GB

POL POL
NL GER
NL GER
UK
BE BE
CZ CZ
LU LU
SK SK
AT AT
HU HUROM ROM

FR
FR CH

CH
HR HR
BA BA
BUL BUL
YU YU
IT
MK MK

AL
TUR
GR GR
CY
MAL

2006 Lenovo

Restriction of Hazardous Substances (RoHS) Directive


In February 2003, the European Union (EU) issued directive 2002/95/EC on the restriction of the
use of certain hazardous substances in electrical and electronic equipment on the EU market
beginning July 1, 2006. The Restriction of Hazardous Substances (RoHS) Directive requires
producers of electrical and electronic equipment to eliminate the use of six environmentallysensitive substances: lead, mercury, cadmium, hexavalent chromium, and the use of
polybrominated biphenyls (PBB) and polybrominated diphenylethers (PBDE) flame retardants. The
purpose is to eliminate the potential risks associated with electronic waste, so this legislation affects
the content and disposal requirements for electronic products.
Most IT hardware is included in scope of the directive: PCs, printers, servers, storage, and options.
Products (and their components) must comply.
Lenovo PC products comply with the RoHS directive across its product lines worldwide.
See www.environment-agency.gov.uk/netregs/legislation/380525/477158/?lang=_e for more
information.

PC Architecture (TXW102)

February 2007

30

Topic 1 - Introduction to Computer Hardware Architecture

A similar directive from the European Union is the Waste Electrical and Electronic Equipment (WEEE)
Directive. WEEE encourages and sets criteria for the collection, treatment, recycling and recovery of
electrical and electronic waste. WEEE requires producers to ensure that equipment they put on the
market in the EU after August 13, 2005 is marked with the crossed-out wheeled bin symbol, the
producers name, and indication that the equipment was put on the market after August 13, 2005.
Lenovo PC products comply with the WEEE Directive requirements.

WEEE-Compliant Symbol

Example of RoHS Labeling on a PCI Express Adapter

PC Architecture (TXW102)

February 2007

31

Topic 1 - Introduction to Computer Hardware Architecture

Industry Standards:
Intel High Definition Audio (Intel HD Audio)
Next-generation architecture (after AC 97) for implementing audio, modem,
and communications functionality

Immersive home-theater-quality sound experience including Dolby 7.1


audio capability

Up to eight channels at 192 kHz with 32-bit quality


Multi-streaming capabilities to send two or more different audio streams to
different places at the same time

Supported with Intel ICH6/ICH7/ICH8 family


- ICH6/ICH7 integrates both AC 97 and HD Audio to facilitate transition
- ICH8 only integrates HD audio (not AC '97)
- Only AC 97 or Intel HD Audio can be used at one time
Game audio

Dolby
Digital

Intel HD Audio
supports multiple
audio streams at
the same time.

Chat audio

2006 Lenovo

Intel High Definition Audio (Intel HD Audio)


Intel High Definition (HD) Audio is an evolutionary technology that replaces AC 97. This next
generation architecture for implementing audio, modem, and communications functionality was
developed to enhance the overall user PC audio experience and to improve stability. Intel HD
Audio facilitates exciting audio usage models while providing audio quality that can deliver
consumer electronics levels of audio experience. The Intel HD Audio specification v1.0 was
released in June 2004.
AC 97
Single stream (in and out)

Intel High Definition Audio


Up to 15 input and 15 output streams at one
time and up to 16 channels per stream

6 channels with 20-bit output, 96 kHz stereo max


12 Mb/s max
Fixed bandwidth assignment
AC 97 DMAs: dedicated function assignment
Codec enumeration at boot time (BIOS)
Codec configuration limitation
12 MHz clock provided by primary codec
Driver software developed by audio codec
supplier

PC Architecture (TXW102)

8 channels with 32-bit output, 192 kHz multichannel


48 Mb/s (SDO), 24 Mb/s (SDI)
Dynamic bandwidth assignment
DMAs: dynamic function assignment
Codec enumeration done by software
(bus driver)
No codec configuration limitation
24 MHz clock provided by the ICH6
OS native bus driver and Independent
Hardware Vendor value-added function driver

February 2007

32

Topic 1 - Introduction to Computer Hardware Architecture

Support for Intel HD Audio is found in the ICH6, ICH7, and ICH8 families. The ICH6 and
ICH7 integrate both AC 97 and Intel HD Audio to facilitate transition from the older AC 97;
however, only AC 97 or Intel HD Audio can be used at one time. (Either requires an additional
external codec chip; when Intel HD Audio was announced, the older AC 97 chips cost less
money.) The ICH6/ICH7 Intel HD Audio digital link shares pins with the AC 97 link. For
input, the ICH6/ICH7 adds support from an array of microphones that can be used for enhanced
communication capabilities and improved speech recognition. The ICH8 only supports HD
Audio (not AC '97).
Intel HD Audio has support for a multi-channel audio stream, a 32-bit sample depth, and a
sample rate up to 192 kHz.
Intel HD Audio delivers significant improvements over previous-generation integrated audio
and sound cards. Intel HD Audio hardware is capable of delivering the support and sound
quality for up to eight channels at 192 kHz/32-bit quality, while the AC 97 specification can
only support six channels at 48 kHz/20-bit quality. In addition, by providing dedicated system
bandwidth for critical audio functions, Intel HD Audio is architected to prevent the occasional
glitches or pops that other audio solutions can have.
Dolby Laboratories selected Intel HD Audio to bring Dolby-quality surround sound
technologies to the PC, as part of the PC Logo Program that Dolby recently announced. The
combination of these technologies marks an important milestone in delivering quality digital
audio to consumers. Intel HD Audio will be able to support all the Dolby technologies,
including the latest Dolby Pro Logic IIx, which makes it possible to enjoy older stereo content
in 7.1-channel surround sound.
Standardized
Register Interface
(UAA)
OS
Audio
driver
Modem
driver

UAA
bus driver
AC 97
drivers

Intel HD
Audio
registers
AC 97
registers

Modem codec
Audio codec

ICH6
Intel HD
Audio
Controller
AC 97
Cntrl

Telephony codec

Intel HD
Audio Link

AC Link

Only Intel HD Audio or AC 97


may be used at one time

Audio
codec

HDMI codec
Dock

Dock codec

Modem
codec

Intel HD Audio Overview

Intel HD Audio also allows users to play back two different audio tracks, such as a CD and a
DVD simultaneously, which can not be done using current audio solutions. Intel HD Audio
features multi-streaming capabilities that give users the ability to send two or more different
audio streams to different places at the same time, from the same PC.

PC Architecture (TXW102)

February 2007

33

Topic 1 - Introduction to Computer Hardware Architecture

Microsoft has chosen Intel HD Audio as the main architecture for their new Unified Audio
Architecture (UAA), which provides one driver that will support all Intel HD Audio controllers
and codecs. While the Microsoft driver is expected to support basic Intel HD Audio functions,
codec vendors are expected to differentiate their solutions by offering enhanced Intel HD audio
solutions.
Intel HD Audio also enables enhanced voice capture through the use of array microphones,
giving users more accurate speech input. While other audio implementations have limited
support for simple array microphones, Intel HD Audio supports larger array microphones. By
increasing the size of the array microphone, users get incredibly clean input through better noise
cancellation and beam forming. This produces higher-quality input to voice recognition, Voice
over IP (VoIP), and other voice-driven activities.
Intel HD Audio also provides improvements that support better jack retasking. The computer
can sense when a device is plugged into an audio jack, determine what kind of device it is, and
change the port function if the device has been plugged into the wrong port. For example, if a
microphone is plugged into a speaker jack, the computer will recognize the error and can
change the jack to function as a microphone jack. This is an important step in getting audio to a
point where it just works. (Users wont need to worry about getting the right device plugged
into the right audio jack.)
The Intel HD Audio controller supports up to three codecs (such as an audio codec or modem
codec). With three Serial Data In (SDI) and one Serial Data Out (SDO) signals, concurrent
codec transactions on multiple codecs are made possible.
The SDO connects to all codecs and provides a bandwidth of 48 Mb/s. Each of the three SDIs
are typically connected to a codec and have a bandwidth of 24 Mb/s. In addition, the controller
has eight non-dedicated, multipurpose DMA engines (4 input, and 4 output). This allows
potential for full utilization of DMA engines for better performance than the dedicated function
DMA engines found in AC 97. In addition, dynamic allocation of the DMA engines allows
link bandwidth to be managed effectively and enables the support of simultaneous independent
streams. This capability enables new exciting usage models (e.g., listening to music while
playing a multi-player game on the Internet).

Surround

Surround

Center
Right

Left

Dolby Digital
DTS
DVD Video
CD

Subwoofer
With Intel HD Audio, a DVD movie with 5.1 audio can be sent
to a surround sound system in the living room, while you
listen to digital music and surf the Web on the PC.

PC Architecture (TXW102)

February 2007

34

Topic 1 - Introduction to Computer Hardware Architecture

Industry Standards:
Intel Active Management Technology (AMT)
Remote management of networked platforms in any system state (even if
powered off or crashed)

Remote hardware and software asset tracking eliminates manual inventory


tracking

Hardware (requires an Intel Ethernet controller) and firmware solution uses


out-of-band communication for system access

Lenovo offers ThinkCentre desktops with Intel AMT support


Out-of-Band
System
Management

Remote
Troubleshooting
and Recovery

Proactive
Alerting

Remote
Hardware and
Software
Tracking

Non-Volatile
Storage

TamperResistant
Agents

Allows remote
management
of platforms
regardless of
power on/off
state or OS
state

Significantly
reduces deskside visits,
increasing the
efficiency of IT
technical staff

Decreases
downtime and
minimizes
time-to-repair

Eliminates timeconsuming
manual inventory
tracking and
human error,
reducing asset
accounting costs
and increasing
tracking accuracy

Survives
power outages
and system
rebuilds

Prevents users
from removing
critical
inventory,
remote-control,
or virusprotection
agents

2006 Lenovo

Intel Active Management Technology (AMT)


Intel Active Management Technology provides capabilities that make new levels of IT
management possible. This platform-resident hardware (requires an Intel Ethernet controller)
and firmware solution uses out-of-band (OOB) communication for platform access regardless of
the state of the operating system (OS) or platform power. Essentially, the only requirements are
that the platform be corporate network connected and have stand-by power (plugged into an
electrical outlet). Even when a platform is powered down or has a non-operational OS, Intel
Active Management Technology maintains access to and management of the platform. This
"any platform state" access gives corporate IT departments unprecedented power for more
efficient platform management with reduced IT operating costs. A software-only solution does
not provide the same level of management when a system is crashed or powered off. See
www.intel.com/technology/manage/iamt/ for more information.
The ability to communicate with a machine that is powered down has been available for a
decade through use of the "magic packet." What is new with AMT is that the capability to wake
up a machine is contained in a piece of hardware and not just in the operating system, where it
has resided in the past.
Extensive surveys of numerous IT shops laid the groundwork for defining Intel Active
Management Technology. Three of the top IT needs revealed by these surveys are better asset
management, reduced downtime, and minimized desk-side visits. Intel design teams determined
that these issues are best addressed through platform architectural enhancements, resulting in
the following solutions:
OOB System Management Allows remote management of platforms regardless of power
on/off state or OS state

PC Architecture (TXW102)

February 2007

35

Topic 1 - Introduction to Computer Hardware Architecture

Remote Troubleshooting and Recovery Significantly reduces desk-side visits, increasing the
efficiency of IT technical staff
Proactive Alerting Decreases downtime and minimizes time-to-repair
Remote Hardware and Software Tracking Eliminates time-consuming manual inventory tracking
and human error, reducing asset accounting costs and increasing tracking accuracy
Non-Volatile Storage Survives power outages and system rebuilds
Tamper-Resistant Agents Prevents users from removing critical inventory, remote-control, or virusprotection agents
Physical implementation of Intel Active Management Technology utilizes hardware and firmware on
the platform's systemboard. Platform auxiliary power maintains this technology's features even when
the platform is turned off.
1

IT Management Console

System problem
occurs

em
st
y
S

wn
do

Problem alert to IT
Remote reboot from IT
management console

Remote control and diagnostics of


platform using ISV applications
Numbers refer to process steps

Corporate Network

To appreciate the importance of the OOB management feature, consider the case where a network
platform's OS becomes non-operational, as shown in the image above. In the traditional IT
environment, the end user calls the IT department to report the problem. An IT technician is then sent to
assess and fix the problem, hopefully in a single desk-side visit or two. On platforms using this new
technology, problems with a non-operational OS can be quickly and efficiently resolvedpossibly even
avertedwithout any desk-side visits. The following sequence explains how:
Proactive alerting reports platform deficiencies, such as an inoperable OS, to the IT management
console. The proactive alerting feature of this new technology uses OOB communication, which is
not handled by the OS. As a result, the inoperable OS cannot stop the alert from being sent. The IT
management console receives the alert about the inoperable OS, and the console operator knows what
has happened, precluding the need for a desk-side visit.
Remote booting allows the console operator to take control of the crashed platform by remotely
booting it to an IT diagnostic platform. At this point, the console operator uses existing management
software to assess and fix the inoperable OS. This could include reinstallation of the IT-approved OS
and patches from the IT management platform, eliminating the requirement of a desk-side visit.
Remote management uses third-party IT management tools that work through a uniform networkconnected application programming interface (API) provided by Intel Active Management
Technology.

PC Architecture (TXW102)

February 2007

36

Topic 1 - Introduction to Computer Hardware Architecture

The image below shows how a corporate IT department can use Intel Active Management Technology
to inventory platform asset information stored in non-volatile, always-available memory. This alwaysavailable memory allows the IT department to continue tracking platform inventory in any system state.
Because it is non-volatile and tamper resistant, asset inventory information cannot be removed,
intentionally or inadvertently, from platforms, and it survives OS rebuilds. This allows more efficient
control and use of network platforms, reducing the number of unidentifiable platforms on the intranet.
1

IT Management Console

2
1
2
1

ISV application polls


platforms for inventory

Platforms report inventory to


IT Management Console

Polling and reporting work


in any system state

Numbers refer to process steps

Corporate Network

PC Architecture (TXW102)

February 2007

37

Topic 1 - Introduction to Computer Hardware Architecture

Industry Standards:
Intel Viiv Technology
Intel brand name for PCs designed for digital entertainment in the home
PCs that meet specific criteria (hardware and software) receive this
branding

Features:
- Instant on/off
- Simple navigation via remote control
- Smaller and quieter systems

Over 100 companies offer content services and products with Viiv logo
Lenovo does not have any products utilizing this branding

2006 Lenovo

Intel Viiv Technology


In January 2006, Intel announced the brand of Intel Viiv technology. Notebook and desktop PCs
that meet specific criteria can receive this branding; the criteria includes specific hardware and
software requirements.
Intel Viiv technology-based PCs feature:
Consumer electronics-like features for simplified entertainment
Instant on/off
Simple navigation to online services with a remote control
Smaller and quieter systems
Performance for high definition entertainment
Intel's latest dual-core processors
Support for up to 7.1 surround sound
Support for high definition video including content downloaded from the Internet
More than 100 companies offer content services and software that have been verified to work with
Intel Viiv PCs. Look for the Enjoy with logo to identify services and applications that have been
verified to work with Intel Viiv technology.
See intel.com/products/viiv for more information.

PC Architecture (TXW102)

February 2007

38

Topic 1 - Introduction to Computer Hardware Architecture

Industry Standards:
Intel vPro Technology
Intel brand name for management of corporate desktop PCs
Desktop PCs that meet specific criteria (hardware and software)
receive this branding

Features:
- Diagnose, reboot, and repair desktops down-the-wire
- Agent presence checking
- Remote asset inventory
- Security virtualization

Select Lenovo ThinkCentre desktops are vPro-compliant

2006 Lenovo

Intel vPro Technology


In April 2006, Intel announced the brand of Intel vPro Technology. Desktop PCs (does not
apply to notebooks) that meet specific criteria can receive this branding; the criteria includes
specific hardware and software requirements.
Desktop PCs with Intel vPro technology include built-in, hardware-based capabilities that can
allow remote management, maintenance, and update of PCs that have traditionally been
inaccessible from the IT management console. A console can remotely communicate with these
PCs, even if system power is off, the operating system is inoperative, or software agents are not
yet installed.
Intel vPros primary feature is the latest version of Intel Active Management Technology
(IAMT version 2) and Intel Virtualization Technology. It allows the implementation of virtual
software appliances to enhance security and reliability of corporate PCs; this includes security
virtualization which enables security software to run as a software appliance in a single virtual
machine partition isolated from the operating system (Lenovo implements this with its
ThinkVantage Virtualization Module for Antidote Delivery Manager).
See intel.com/vpro for more information.

PC Architecture (TXW102)

February 2007

39

Topic 1 - Introduction to Computer Hardware Architecture

Benchmarks
Understand benchmark objective:
either application throughput or
subsystem performance

Examples include:
- BAPCo

SYSmark 2004 SE

- MobileMark 2005
- 3DMark05 and 3DMark06
Lenovo ThinkCentre A60

- SPEC CPU2000

PC performance
doubles every
two years.

2006 Lenovo

Benchmarks
The following is a short list of benchmarks and the systems they measure.
Overall performance:
SYSmark 2004 SE - SYSmark includes office productivity and Internet content creation
benchmark tests. The two scores are combined and given a weighted average to produce an
overall performance rating. Both SYSmark tests derive scores by using real-world
applications to run a preset script of user-driven workloads and usage models developed by
application experts.
The SYSmark 2004 SE Internet Content Creation test is organized as scenarios that are
designed to simulate an Internet content creators day. This benchmark incorporates such
applications as Adobe Photoshop 7.01, Discreet 3ds max 5.0, and Macromedia
Dreamweaver MX.
The SYSmark 2004 SE Office Productivity test follows ICCs blueprint by mimicking the
usage patterns of todays desktop and mobile business users, including the concurrent
execution of multiple programs. Applications such as Adobe Acrobat 5.0.5, McAfee
VirusScan 7.0, and the Microsoft Office suite are used. Each SYSmark test measures the
response time of the application to user input. Both scores are combined using a geometric
mean to get an overall score.

PC Architecture (TXW102)

February 2007

40

Topic 1 - Introduction to Computer Hardware Architecture

The SYSmark benchmarks are created by BAPCo (Business Applications Performance


Corporation), which is a nonprofit corporation founded in May 1991 to create objective
performance benchmarks that are representative of the typical business environment. For
notebook systems, MobileMark 2005 is a benchmark created by BAPCo that measures both
performance and battery life at the same time using popular applications. Contact
www.bapco.com for more information.
MobileMark 2005 MobileMark 2005 is the latest version of the premier notebook battery
life and performance under battery life metric based on real world applications.
Graphics performance
3DMark03, 3DMark05, 3DMark06, and 3DMarkMobile06 are benchmark tests that run
through different scenes using various DirectX or OpenGL calls to derive a score reflecting
the graphics hardware and driver performance. See www.futuremark.com/products/ for
more information.
Battery life
Business Winstone BatteryMark (BWS BatteryMark) 2004 measures the battery life of
notebook computers, providing users with a good idea of how long a notebook battery will
hold up under normal use. This benchmark uses the same workload as in Business
Winstone 2004.
Notable benchmark organizations
In 1988, the Transaction Processing Council (TPC) was formed to fulfill the need for
transaction processing benchmarks that emulate the workloads found on database servers.
The council includes representatives from a cross-section of 45 hardware and software
companies that meet to establish benchmark content. A primary goal of the council is to
provide objective and verifiable performance data to the industry. Visit www.tpc.org for
more information.
The Standard Performance Evaluation Corporation (SPEC) establishes, maintains, and
endorses a standardized set of relevant benchmarks and metrics for performance evaluation.
Contact www.specbench.org for more information.
SPEC CPU2000 - Introduced in late 2000, SPEC (Standard Performance Evaluation Corp.)
CPU2000 is a workstation application-based benchmark program that can be used across
several versions of Microsoft Windows NT, Windows 2000, and Unix. It consists of the
two benchmark suites listed below. Both measure the real-world performance of a
computers processor, memory architecture, and compiler. They replace CPUmark and
FPUmark.:
SPECINT2000 measures computation-intensive integer performance
SPECFP2000 measures computation-intensive floating point performance.

PC Architecture (TXW102)

February 2007

41

Topic 1 - Introduction to Computer Hardware Architecture

Security Issues:
Client Security Solution
ThinkVantage software (Client Security Solution) offers
unique advantages

Robust security solution


Hardware and software combination

ThinkPad security chip


2006 Lenovo

Client Security Solution


ThinkVantage Client Security Solution is a unique hardwaresoftware combination helps protect your company information,
including vital security information like passwords, encryption
keys and electronic credentials, while helping to guard against
unauthorized user access to data.
This level of security is critical for both desktop and notebook
systems. In fact, you cannot get a higher level of security on
select ThinkPad notebooks or ThinkCentre desktops as a
standard feature on a PC from any other manufacturer.

ThinkCentre security chip

ThinkPad security chip

PC Architecture (TXW102)

February 2007

42

Topic 1 - Introduction to Computer Hardware Architecture

New Features
Active Directory support for seamless configuration and management support
Completely new and easier to use password manager with broader browser support, autorecognize/fill, and per-site security policies
Multi-factor support and policy manager for improved security
Updated fingerprint reader software with integrated tutorial
ThinkVantage Client Security Solution helps turn your computer into a highly protected vault.
Available preloaded or by download for all ThinkPad notebooks and ThinkCentre desktops, it
provides advanced technology for user authentication plus enhanced security for wired and
wireless networking.
To further enhance security, select ThinkPad and ThinkCentre systems also include an
embedded Trusted Platform Module (TPM). ThinkVantage Client Security Solution works in
conjunction with this chip to manage encryption keys and processes.
The TPM itself is isolated from the operating system using patented tamper-resistant
technology. To initialize the TPM, users or administrators simply download the Client Security
Solution and run a single setup wizard to install the software and create a set of master
encryption keys.
With new Active Directory integration, IT administrators gain significant flexibility and control
over how CSS works in the customer environment. From logon to password manager, IT gets
tremendous flexibility to deploy and enforce policies and customization options.
Security made easy
The Client Security Software wizard helps simplify installation and setup and includes a set of
predefined security settings, allowing you to select the best suited to your needs. For enterprise
customers, script-driven silent installations can facilitate large deployments and are
complemented by the ability to push policies to clients from a central server.
Simplify passwords: The Client Security Password Manager replaces multiple passwords with
one easy-to-remember password or fingerprint (requires an optional fingerprint reader
device), thereby reducing password reset calls to the help desk.
Reset forgotten passwords: Users can pre-configure personal questions for this purpose. If the
user later forgets his Power On password, Hard Drive password or Windows password, he
can click a button, answer his pre-configured personal questions and CSS will reset his
forgotten password, require a new password, and enable the user to log on.
Simplify encryption: "On-the-fly" file and folder encryption saves you time by automatically
encrypting files and helping to protect data from unauthorized access.

PC Architecture (TXW102)

February 2007

43

Topic 1 - Introduction to Computer Hardware Architecture

Flexible security
Client Security Solution supports the use of a password, a passphrase or a fingerprint as
authentication credentials for access to protected data. Where central management is an
important requirement, Client Security Solution can be integrated into a Windows Domain
Services management model. If very sensitive data must be protected on a PC, the customer can
create policies that force multi-factor authentication and can integrate Client Security Solution
with best-of-breed data encryption solutions.
Client Security Solution includes Utimaco Private Disk Personal Edition so users can
automatically lock valuable data with file and folder encryption. The size of the "disk volume"
to be encrypted can be set by IT staff or controlled by the user. Only verified users can unlock
the encrypted data, so sensitive information is better protected should a computer be stolen. For
users interested in protecting all the data on their computer, full hard drive encryption is also
available from Lenovo with Utimaco SafeGuard Easy. (Availability varies by country.)
Enhanced security for both wired and wireless
Client Security Solution provides enhanced security for both wired and wireless networks. In
both cases, the Client Security Solution and TPM ensure data confidentiality and availability by
providing a hardware- and software-based architecture to provide better protection for sensitive
keys, identity information and confidential data. Further, for wireless networks, the TPM
hardware provides enhanced authentication and session confidentiality by concealing
authentication credentials for industry-standard 802.1x protocols and Cisco LEAP.
Advanced protection for your digital identity
By requiring an advanced means of authentication for your systemwhat you know (a
passcode), what you have (using optional solutions such as a token) or who you are (biometric
scan, such as a fingerprint)you can help keep your digital identity and data more secure. The
Client Security Solution and TPM store and enable user authentication through integrated
hardware rather than in the software or memory, both of which are more vulnerable to
unauthorized users. It also provides additional protection for digital IDs used for e-business
transactions and network access.
The ThinkVantage Client Security Solution is certified RSA SecurID-ready for remote access
protection, and can function as an RSA SecurID authenticator. As a result, mobile users can
employ cost-efficient SecurID software tokens and carry one less piece of equipment on the
roadwithout sacrificing security or budget.

PC Architecture (TXW102)

February 2007

44

Topic 1 - Introduction to Computer Hardware Architecture

ThinkCentre Security Chip


(2003 to 2004)

Security chip was a separate module from Atmel that plugged into
systemboard

ThinkCentre Security Chip


(2004 to 2006)

Security chip was integrated in Super I/O on systemboard as the National


Semiconductor PC8375T or Winbond PC8375S (TPM 1.2-compliant) Super I/O for
security chip, Asset ID, serial, parallel, diskette, keyboard, mouse, auto-thermal
controls, hardware event log

PC Architecture (TXW102)

February 2007

45

Topic 1 - Introduction to Computer Hardware Architecture

ThinkPad Security Chip


(2002 to 2003)

Security chip was a separate module from Atmel mainly used in ThinkPad T3x
systems that plugged into systemboard

ThinkPad Security Chip


(2003 to 2004)

Security chip was soldered to systemboard and was a module from Atmel
mainly used in select ThinkPad notebooks such as the T40/T41/T42 systems

ThinkPad Security Chip


(2005)

Security chip was integrated in Super I/O on systemboard as the National


Semiconductor PC8394T Super I/O for security chip, Asset ID, hardware event log,
parallel, infrared, keyboard, mouse [docking station ports (serial, diskette)]

PC Architecture (TXW102)

February 2007

46

Topic 1 - Introduction to Computer Hardware Architecture

ThinkPad Security Chip


(2005 to 2006)

Security chip was soldered to systemboard and was a module from Atmel
mainly used in select ThinkPad notebooks such as the T60/X60/Z60 systems

PC Architecture (TXW102)

February 2007

47

Topic 1 - Introduction to Computer Hardware Architecture

Trusted Platform Module


A Trusted Platform Module (TPM) is a special-purpose integrated circuit (IC) built into a variety of
platforms to enable strong user authentication and machine attestationessential to prevent
inappropriate access to confidential and sensitive information and to protect against compromised
networks. Trusted Platform Modules utilize open standards and technologies to ensure interoperability
of diverse products in mixed-vendor environments.
The TPM is based on specifications developed by the Trusted Computing Group (TCG). The TCG is
an industry standards group formed to develop, define, and promote open standards for trusted
computing and security technologies, including hardware building blocks and software interfaces,
across multiple platforms, peripherals, and devices. Members include Microsoft, Intel, Dell, HP, IBM,
and Lenovo. Current TPM implementations are based on the TCG 1.1 specification. TPM
implementations based on the next-generation version, TCG 1.2, started shipping in 2005. Microsoft's
Vista due in 2006 will support both TPM 1.1 and 1.2. See www.trustedcomputinggroup.org for
more information.
How a Trusted Platform Module (TPM) Secures A System

The logon user name and


password are stored in the
TPM chip.

User ID
Password

TPM
chip

When files and folders are encrypted, they can


be decrypted only by the person who has the
authentication data the TPM chip requires.

abcdefg
Disk

*******

Encrypted Files
Smart-card
reader
Fingerprint and smart-card readers
can be used in addition to the TPM for
an additional security layer.
Fingerprint reader

Anyone trying to hack into your system


will not be able to read any of the
encrypted files or folders.

TPM
chip
TPM chip

Encrypted files

If a disk is removed, the encrypted data is safe


because it can not be decrypted without being
authenticated through the TPM.

PC Architecture (TXW102)

February 2007

48

Topic 1 - Introduction to Computer Hardware Architecture

User Authentication without TPM

User Authentication with TPM

Inadequate user ID and password protection


makes "spoofing" very easy

Strong protections eliminate "spoofing"; verifies


integrity of user login credentials

Multiple login IDs and passwords cause users to On-chip, protected storage of secrets reduces
be careless; store secrets without protecting
user burden; enables secure single sign-on;
them; use weak protections
ensures strong protections
Storage of IDs and passwords in easily copied
files; use of one set of secrets for access to all
systems

Secure storage of IDs and passwords; multiple


log-in secrets secured by the TPM

Platform Attestation without TPM

Platform Attestation with TPM

Easy to change settings and parameters for


unauthorized access and malicious damage

Secure access prevents unauthorized access;


secure hash comparison validates settings

Altered settings allow inappropriate access to


valued networks and sensitive data

Validated settings ensure system integrity and


prevent inappropriate access

Untrustworthy systems result in unreliable and


untrustworthy practices

Trustworthy systems result in reliable and


trustworthy practice; reduce support expenses

PC Architecture (TXW102)

February 2007

49

Topic 1 - Introduction to Computer Hardware Architecture

Security Issues:
Integrated Fingerprint Reader
Fingerprint image can replace:

Fingerprint software is used as:

- Power-on password

- An interface to the fingerprint reader to


register and manage user fingerprints

- Supervisor password

- An interface to Client Security


Software as a means of authentication

- Hard drive password


- Windows password

- A means to bypass BIOS and


Windows logon passwords

Easy and reliable to use


- User slides finger over reader
- Reader takes multiple snapshots of finger

- A replacement for Windows 2000 and


Windows XP logon passwords

- Snapshots are assembled into single


image

More productive and secure than


passwords, key fobs, and smart cards
- One swipe at power-on and a PC comes
up to the Windows desktop
- Impossible to lose as the fingerprint
reader is integrated into the system

Integrated fingerprint reader is included


in select ThinkPad notebooks.
2006 Lenovo

Integrated Fingerprint Reader


The integrated fingerprint reader is a new security device introduced on select ThinkPad notebook
models (first introduced on the ThinkPad T42 in October 2004), the Preferred Pro USB Fingerprint
Reader Keyboard, and USB Fingerprint Reader. The device allows users to register their
fingerprints on these systems and then use the reader as an authentication device to replace BIOS
and Windows passwords. The integrated fingerprint reader can also be used as an authentication
device with the security chip included on select ThinkPad and ThinkCentre systems.
The integrated fingerprint reader provides a convenient means for authenticating to a Windows
system, eliminating the need for users to remember multiple passwords and thereby reducing the
number of support calls for forgotten passwords. When used with Client Security Solution with
Password Manager, the fingerprint reader can be used as a means for authenticating to secure
operations. Fingerprint technology provides the security of unique user identification, and the
integrated design means no inconvenient external devices are needed.
The integrated fingerprint reader also comes with a special utility called Fingerprint Software. This
software allows you to enroll fingerprints and adjust security settings related to the fingerprint
sensor.

PC Architecture (TXW102)

February 2007

50

Topic 1 - Introduction to Computer Hardware Architecture

Fingerprint Software Screen

The format of the reader is called a slide sensor. In order to get a reading, you must slide, drag, or
swipe your finger across the reader. The slide sensor takes sequential snapshots of your finger as it
slides over the surface. It then stitches those snapshots together to form a fingerprint image that can
be as large as or even larger than the image taken by a contact sensor.
The advantages of a slide over a contact are as follows:
The slide can make a larger image of the finger being read. This means the matcher software has
more data to analyze and is therefore less likely to make a mistake.
The slide is only 20% of the size of a contact sensor. The much smaller footprint of the slide
sensor compared to the contact sensor is a key feature in terms of ergonomic and engineering
considerations around placement of the device.

PC Architecture (TXW102)

February 2007

51

Topic 1 - Introduction to Computer Hardware Architecture

An integrated fingerprint reader is used on the optional


Preferred Pro USB Fingerprint Reader Keyboard (part number 73P4730)

An integrated fingerprint reader is supported on select Lenovo


systems via the optional USB Fingerprint Reader (part number 41U3167)

PC Architecture (TXW102)

February 2007

52

Topic 1 - Introduction to Computer Hardware Architecture

Ways to Use the Integrated Fingerprint Reader


First, you can enroll fingers for use in pre-boot authentication. That means you can configure your
system to demand a fingerprint when it is turned on, provided that you have configured your system
with one or more BIOS passwords. The fingerprint reader subsystem has the ability to securely
remember these passwords and can provide them to BIOS on your behalf when you correctly
authenticate with a known fingerprint. The reader has the ability to store and recall a power-on
password and up to three different hard drive passwords (one each for up to three drives). When
prompted, if you provide a fingerprint that is matched, then the fingerprint software will release the
requested passwords (provided they have been previously stored) to BIOS. In addition, there is a
configuration option that will enable this same fingerprint to be used to log you into your Windows
account as well; one swipe at power-on and your PC boots up all the way to your desktop.
Second, select ThinkPad models with the integrated fingerprint reader have a trusted platform module
as a standard feature. Therefore, a user can use the reader as another means of authentication for
existing Client Security Solution applications such as Password Manager, Windows password
replacement, Utimaco Private Disk Personal Edition, and certificate protection.
Third, you can use the reader as a way to replace your Windows 2000 and Windows XP logon
passwords. The IBM fingerprint software has the capability to determine who you are so it can select
your user ID and provide your password based on the fingerprint provided; you do not need to enter or
select your user ID at the login screen because the application does it for you. This feature is
independent of Client Security Solution (CSS) so there is no protection by the security chip. For best
security results for users running CSS, implement Windows password replacement through CSS
policy.

Fingerprint Hardware

ThinkPad BIOS

Trusted Platform
Module
Fingerprint Sensor
ThinkPad T42

PC Architecture (TXW102)

February 2007

53

Topic 1 - Introduction to Computer Hardware Architecture

Summary:
Computer Hardware Architecture
Notebooks, desktops, PC Blades, and
servers are different types of computers
with unique features.

Each computer subsystem is connected


by buses or links.

Wintel-based PCs adhere to industry


standards, and industry benchmarks
provide subsystem and total system
performance comparisons.

Security can be enhanced with the

Lenovo 3000 Small Desktop and Tower

ThinkVantage Client Security Solution


and integrated fingerprint reader.

Lenovo ThinkPad Z61t


2006 Lenovo

PC Architecture (TXW102)

February 2007

54

Topic 1 - Introduction to Computer Hardware Architecture

Computer Measurements
Bit (b): on charge or off charge (0 or 1)
Byte (B): eight bits (a character/number represented by a byte)
Kilobyte (2^10): one thousand bytes (KB): 1,024
Megabyte (2^20): one million bytes (MB): 1,048,576
Gigabyte (2^30): one billion bytes (GB): 1,073,741,824
Terabyte (2^40): one trillion bytes (TB): 1,099,511,627,776
Petabyte (2^50): one quadrillion bytes (PB): 1,125,899,906,842,624
Exabyte (2^60): 1,000 petabytes (EB): 1,152,921,504,606,846,976
Zettabyte (2^70): 1,000 exabytes (ZB): 1,180,591,620,717,411,303,424
Yottabyte (2^80): 1,000 zettabytes (YB): 1,208,925,819,614,629,174,706,176
Millisecond: one thousandth of a second (ms): 1/1,000
Microsecond: one millionth of a second (us): 1/1,000,000
Nanosecond: one billionth of a second (ns): 1/1,000,000,000
Picosecond: one trillionth of a second (ps): 1/1,000,000,000,000
Megahertz: millions of cycles per second (MHz)
Gigahertz: billions of cycles per second (GHz); 1GHz=1,000 MHz

PC Architecture (TXW102)

February 2007

55

Topic 1 - Introduction to Computer Hardware Architecture

Clock Cycle Times


The amount of time for a single clock cycle:
12 MHz = 83.3 ns
33 MHz = 30 ns
50 MHz = 20 ns
66 MHz = 15 ns
100 MHz = 10 ns
133 MHz = 7.5 ns
150 MHz = 6.6 ns
166 MHz = 5.9 ns
200 MHz = 5.0 ns
233 MHz = 4.3 ns
266 MHz = 3.75 ns
300 MHz = 3.3 ns
400 MHz = 2.5 ns
533 MHz = 1.875 ns
800 MHz = 1.125 ns
Cycle Speed Formula
1 frequency (in hertz) = speed of one bus cycle in seconds (multiply by ten to the power of nine
to convert to nanoseconds). For example:
1 10 million = 10 power -7 = 100 nanoseconds
Calculate Date Transfer Rates
Frequency (in MHz) data width (in bits) = maximum data bus capacity in Mb/s ( 8 to convert to
MB/s). For example:
10 MHz 32-bits = 320 Mb/s or 40 MB/s

PC Architecture (TXW102)

February 2007

56

Topic 1 - Introduction to Computer Hardware Architecture

Speeds

bps
KB/s
Mb/s
MB/s
Gb/s
GB/s

=
=
=
=
=
=

bits per second


kilobytes per second
megabits per second
megabytes per second
gigabits per second
gigabytes per second

Some magazines (like PC Magazine) measure in megabytes (1,024 bytes squared, or 1,048,576).
For rough conversion, subtract five million from every hundred million bytes. For example, 720
million bytes is 687 megabytes.
Speeds for connectivity are normally measured in bits (bits per second). Speeds for transfers
within a PC (like across PC buses) are normally measured in bytes (bytes per second).
5 MB/s

AT Bus (ISA)

132 MB/s

33 MHz at 32-bit (PCI bus)

528 MB/s

66 MHz at 64-bit

800 MB/s

100 MHz at 64-bit

1,064 MB/s

133 MHz at 64-bit (Pentium II 266/66MHz to 1/2 speed L2 cache)

1,200 MB/s

150 MHz at 64-bit

1,600 MB/s

200 MHz at 64-bit (or 128-bit bus at 100MHz)

1,600 MB/s

800 MHz at 16-bit (rambus DRAM)

3,200 MB/s

Pentium 4 system bus at 400 MHz at 64-bit

3,200 MB/s

400 MHz at 64-bit (Xeon backside bus to full speed L2 cache)

4,200 MB/s

Pentium 4 system bus at 533 MHz at 64-bit

6,400 MB/s

200 MHz at 256-bit

6,400 MB/s

Pentium 4 system bus at 800 MHz at 64-bit

For each 20 degrees Celsius-increase in operating temperature, electronic component life drops in
half.
A billion bits is equivalent to 62,500 double-spaced typewriter pages--enough paper to stack 21
feet high.
0.15 micron is 1/600 the width of one human hair.
Logic circuits are used to process information; memory circuits store information.

PC Architecture (TXW102)

February 2007

57

Topic 1 - Introduction to Computer Hardware Architecture

Network Speeds
Data Rate

Average Usable
Throughput

512 Kb/s-6 Mb/s

150 Kb/s-2 Mb/s

Phone-line (HPNA 2.0) and 802.11b

10 Mb/s

5 Mb/s

Power-line (HomePlug)

14 Mb/s

5 Mb/s

802.11a & 802.11g

54 Mb/s

25 Mb/s

Pre-802.11n

108 Mb/s

50 Mb/s

Fast Ethernet

100 Mb/s

60 Mb/s

1,000 Mb/s

600 Mb/s

Home Networking Technologies


Cable or DSL Internet connection

Gigabit Ethernet

Average Required
Throughput

Applications
Video
Uncompressed full-motion video stream

1,000-2,000 Mb/s

Compressed HDTV

18 Mb/s

Compressed standard / extended definition TV stream

7-15 Mb/s

Audio
Radio-quality MP3 or WMA stream

64 Kb/s

CD-quality MP3 or WMA stream

128 Kb/s

Dolby AC-3 stream

640 Kb/s

Other
VolP traffic

64 Kb/s

Typical broadband surfing traffic

1-256 Kb/s

E-mail traffic

30-100 Kb/s

Broadband Types
Business cable

Business DSL

Fractional T1

Wired MAN

Wireless MAN

2-5 Mb/s
384-768 Kb/s

144 Kb/s-6 Mb/s


144 Kb/s-1.5 Mb/s

384 Kb/s-1.5 Mb/s


symmetrical

10 Mb/s-1 Gb/s
symmetrical

256 Kb/s-100 Mb/s


symmetrical

Pros

Inexpensive

Inexpensive

Reliable

Fast, costeffective

No cabling, instant
installation

Cons

Shared media,
limited bandwidth

Limited bandwidth

Expensive

Limited
availability

Limited availability

Speed
(download/
upload)

PC Architecture (TXW102)

February 2007

58

Topic 1 - Introduction to Computer Hardware Architecture

Review Quiz
Objective 1
1. What type of system would market its wireless capability, weight, and docking station support?
a. Notebook
b. Desktop
c. Server
d. Information appliance
2. What type of system would most likely market its systems management support, chipset, and
graphics performance?
a. Notebook
b. Desktop
c. PC Blade
d. Information appliance
3. The ThinkPad brand name is associated with what type of computer system?
a. PDA
b. Notebook
c. Desktop
d. Server
4. Rescue and Recovery, Access Connections, and Productivity Center are from what strategic
Lenovo offering?
a. ThinkVantage Technologies
b. ThinkVision monitors
c. ImageUltra Builder
d. System Migration Assistant
Objective 2
5. A device driver is an interface between what subsystems?
a. The applications and BIOS
b. Hardware and the operating system
c. The operating system and the BIOS
d. An API and the standard hardware
6. What layer is between the physical hardware and the operating system and allows for software
compatibility?
a. Applications programming interface
b. Device driver
c. Operating system
d. BIOS

PC Architecture (TXW102)

February 2007

59

Topic 1 - Introduction to Computer Hardware Architecture

7. What circuitry controls the methods, manner, and speed in which a subsystem is accessed?
a. Controller
b. Keyboard
c. VLSI logic
d. Device driver
8. Most transfers of data between subsystems involve which buses?
a. Data bus
b. Data and address bus
c. Data, address, and control bus
d. Data, address, control, and tag bus
Objective 3
9. What is the name of the architecture for implementing audio, modem, and communications
functionality after AC 97?
a. Intel High Definition Audio (Intel HD Audio)
b. I/O Controller Hub 6 (ICH6)
c. Unified Audio Architecture (UAA)
d. Dolby Digital
10. What is an important characteristic of a performance benchmark?
a. Understanding if it measures a subsystem or application throughput
b. The benchmark needs to incorporate Java
c. Benchmark must include MPEG-2 encoding
d. The PCI bus must be enabled
11. What is the Intel brand name for PCs designed for digital entertainment in the home?
a. Intel High Definition Audio (Intel HD Audio)
b. Intel Viiv Technology
c. Intel vPro Technology
d. Intel Advanced Digital Media Boost
12. What is the Intel brand name for management of corporate desktop PCs?
a. Intel High Definition Audio (Intel HD Audio)
b. Intel Viiv Technology
c. Intel vPro Technology
d. Intel Advanced Digital Media Boost

PC Architecture (TXW102)

February 2007

60

Topic 1 - Introduction to Computer Hardware Architecture

Objective 4
13. What single device can be used to replace the power-on password, supervisor password, hard
drive password, and Windows password?
a. Optical drive
b. Digital certificate
c. Integrated fingerprint reader
d. Encryption key

PC Architecture (TXW102)

February 2007

61

Topic 1 - Introduction to Computer Hardware Architecture

Answers
Objective 1
1. A
2. B
3. B
4. A
Objective 2
5. B
6. D
7. A
8. C
Objective 3
9. A
10. A
11. B
12. C
Objective 4
13. C

PC Architecture (TXW102)

February 2007

62

You might also like