Nof S.Y. Springer Handbook of Automation

Подождите немного. Документ загружается.

Education and Qualiﬁcation for Control and Automation 44.7 The Scholarship of Teaching and Learning 775

sider a career in control was the workshop organized

for girls at the International Symposium on Intelligent

Control/Mediterranean Conference on Control and Au-

tomation (ISIC/MED) conference [44.48].

The purpose of NSF/CSS workshops is to inspire

interest from youth towards studies in control systems

and to assist high-school teachers in promoting the dis-

cipline of control systems among their students. It is

composed of several short but effective presentations,

as listed in Sect.44.5.3, on various problems from the

real world that have been solved by using control en-

gineering methods, techniques, and technologies. The

workshops bring together all-star teams of some of the

most eminentsenior control researchers and some of the

most prominent younger researchers involved in new

technologies to present control systems as an exciting

and intellectually stimulating ﬁeld. The attractiveness

and excitement of choosing a career in controlengineer-

ing has been addressed at workshops. Live interaction

between thepresenters and the audiencehas beenan im-

portant feature of the workshops. The workshops have

become very popular, with the last one in San Diego

bringing over 650 students and teachers. Additional in-

formation on the workshops can be found in [44.49,50].

44.6 Integrating Scholarship, Teaching, and Learning

The integration of scholarship, teaching, and learn-

ing into the classroom can be subdivided in many

ways. An important subdivision is horizontal inte-

gration versus vertical integration [44.51]. Horizontal

integration deals with institutional integration, drawing

perspectives from across the institution, whereas verti-

cal integration is programmatic, drawing perspectives

from throughout a speciﬁc school, department, or pro-

gram. Both are important for success.

To increase horizontal integration, faculty from

different disciplines work together in teaching a sin-

gle course. Faculty in different disciplines often work

on similar topics and each one can provide his or

her particular insights for understanding the mater-

ial and assimilating the information. Their combined

efforts result in a variety of advantages. First, they

develop cross-disciplinary ties which often boost col-

laboration. Second, students learn a topic from multiple

perspectives, which increases their potential ability to

understand the topics, as they are more likely to hear at

least one explanation that they will understand, as well

as increasing their interactions with a wider variety of

faculty members.

Vertical integration incorporates students and re-

searchers at different levels in the teaching activity, that

is, there is involvement of senior high-school students,

undergraduates, graduates, and postdoctoral students.

Often a student is more likely to discuss various ques-

tions with someone near to his or her educational level,

and these educationally close people can often iden-

tify more easily the causes of difﬁculty. The Stochastic

Adaptive Control Group (SACG) research group at the

University of Kansas has successfully implemented the

approach [44.46,49,52].

44.7 The Scholarship of Teaching and Learning

No modern educational discussion would be complete

without some discussion of the scholarship of teach-

ing and learning (SoTL). While the wording of SoTL

is similar to that of Sect. 44.6, there is a careful distinc-

tion. Integrating scholarship, teaching, and learning is

about using research (scholarship) in the classroom to

enhance teaching and learning. SoTL is about analyzing

and reﬂecting on teaching and learning using the same

scholarly processes (literature review, testing, data col-

lection, statistics, as well as more qualitative measures

such as systematically coded interviews) used in other

scholarly work. Illinois State University deﬁnes SoTL

to be “systematic reﬂection on teaching and learning

made public” [44.53].

What makes aspects of SoTL interesting for control

educators is that the process of systematically reﬂecting

can itself be a type of control problem. We learn about

our students each time weteach and weadapt the course

and methods of teaching to this particular class. It can

be considered a stochastic system; stochastic because

there is a lot of noise in the system, but typically also

having some degree of consistency. Sometimes referred

to as action research, the process involves an itera-

tive cycle of planning, action, and fact-ﬁnding about

Part E 44.7

776 Part E Automation Management

the result of the action [44.54]. With each new set

of facts, a new plan is formed, upon which action is

taken and new facts emerge. The process then begins

anew. Over time, a lesson is reﬁned, with the best ap-

proaches and attributes of the lesson being used, while

the weakest elements are systematically removed or

replaced. Boyer refers to this as “scholarship in teach-

ing” [44.55].

Education should be regarded as a stochastic pro-

cess that changes over time, a process with several

components such as vision, design, data collection, and

data analysis. As instructors, we shouldcollect informa-

tion, build a portfolio, analyze our reports and data after

every class, and want to do better each time. We should

want to apply the same rigor to our teaching that we do

to our other scholarly endeavors.

44.8 Conclusions and Emerging Challenges

The control ﬁeld is an exciting ﬁeld. It has a cross-

boundary nature. Many nontraditional disciplines rec-

ognize the power and the need for systems and control

approaches, principles, and technologies. Control edu-

cation of tomorrow is a collaborative effort integrating

scholarship, teaching, and learning. This collabora-

tive effort needs to include K-12 teachers, students,

and scholars. They need to work together as partners

who are learners in the process of control education.

A classroom should be treated as a scientiﬁc laboratory,

actively engaged in SoTL.

Similarly, instructors must learn to integrate their

scholarship into their teaching and learning by building

ties both across a discipline and across an institution. It

is important for control instructors to build bridges with

mathematics, computer science, and science. A future

control engineer has to integrate engineering with com-

putations, communications, mathematics, and science.

This is an extraordinary time for control, with ex-

traordinary opportunities. Several special sessions on

control educations have been organized at major control

conferences, bringing together leading control scholars

from academia and industry. Those important discus-

sions have been well documented [44.16]. Every issue

of the IEEE Control Systems Magazine is either devoted

fully or has an article on control education. The biggest

challenge is to attract young people to engineering and,

in particular, systems and control engineering. It is im-

portant for all control communities to be involved in

outreach programs. It is important to build new bridges

with other disciplines. It is important to focus on good

communication and writing and it is most important to

be passionate and enthusiastic about systems and con-

trol, and pass on this passion and enthusiasm to young

people, in particular women and minorities, encourag-

ing them to pursue and stay in the control profession,

which has so much to offer, both now and as control ex-

pands into more and more new and unique disciplines.

Future engineers must be well prepared for the

complex technical, social, and ethical questions raised

by emerging disciplines and technologies. To be suc-

cessful, students will need to be broadly educated.

Teachers and students need to understand the full range

of problems covering requirement, speciﬁcations, im-

plementation, commissioning, and operation that is

reliable, efﬁcient, and robust. They need to under-

stand that new materials and devices are made possible

through advanced control of manufacturing processes.

They need to recognize that control theories can be used

for achieving breakthroughs in highly diverse settings,

including biomedicine and ﬁnance. As the NSF/CSS

panel report summarizes [44.3]:

... perhaps most important is the continued de-

velopment of individuals who embrace a system

perspective and provide technical leadership in

modeling, analysis, design and testing of complex

engineering systems.

References

44.1 S. Bittanti, M. Gevers (Eds.): On the dawn and

development of control science in the XX-th

century (Special Issue), Eur. J. Control 13,1–81

(2007)

44.2 W.H. Fleming: Future Directions in Control The-

ory. A mathematical Perspective (Society for

Industrial and Applied Mathematics, Philadelphia

1988)

44.3 P. Antsaklis, T. Basar, R. DeCarlo, N.H. McClam-

roch, M. Spong, S. Yurkovich: Report on the NSF/CSS

workshop on new directions in control engineering

education, IEEE Control Syst. Mag. 19, 53–58 (1999)

Part E 44

Education and Qualiﬁcation for Control and Automation References 777

44.4 R.M. Murray: Control in an Information Rich World

(Society for Industrial and Applied Mathematics,

Philadelphia 2003)

44.5 R.M. Murray, K.J. Åström, S.P. Boyd, R.W. Brockett,

G. Stein: Future directions in control in an infor-

mation rich world, IEEE Control Syst. Mag. 23,20–23

(2003)

44.6 E.D. Sontag: Molecular systems biology and control,

Eur. J. Control 11, 396–436 (2005)

44.7 J.M. Bailey, W.M. Haddad: Paradigms, beneﬁts,

and challenges. Drug dosing control in clinical

pharmacology, IEEE Control Syst. Mag. 25, 35–51

(2005)

44.8 S.H. Haas, M.G. Frei, I. Osorio, B. Pasik-Duncan,

J. Radel: EEG ocular artifact removal through

ARMAX model system identiﬁcation using ex-

tended least squares, Commun. Inf. Syst. 3,19–40

(2003)

44.9 I. Osorio, M.G. Frei: Hurst parameter estimation for

epileptic seizure detection, Commun. Inf. Syst. 7,

167–176 (2007)

44.10 I. Osorio, M.G. Frei, S.B. Wilkinson: Real time au-

tomated detection and quantitative analysis of

seizures and short term prediction of clinical onset,

Epilepsia 39, 615–627 (1998)

44.11 B. Pasik-Duncan: Random walk around some

problems in identiﬁcation and stochastic adap-

tive control with applications to ﬁnance. In:

AMS-IMS-SIAM Joint Summer Research Conference

on Mathematics of Finance (2003) pp. 273–

287

44.12 B. Pasik-Duncan: Special issue on the stochas-

tic control methods in ﬁnancial engineering, IEEE

Trans Automatic Control (2004)

44.13 P. Varaiya: Reducing highway congestion: an em-

pirical approach, Eur. J. Control 11, 301–310 (2005)

44.14 P.R. Kumar: New technological vistas for systems

and control: the example of wireless networks.

What does the future hold for the design of wire-

less networks?, IEEE Control Syst. Mag. 21,24038

(2001)

44.15 T.E. Djaferis: Systems and control impact in

a changing world. In: Chinese Control Conference

(CCC) (Hunan, China 2007) pp. 21–28

44.16 B. Pasik-Duncan, R. Patton, K. Schilling, E.F. Ca-

macho: Four focused forums. Math, science and

technology in control engineering education, IEEE

Control Syst. Mag. 26, 93–98 (2006)

44.17 G. Clough: The Engineer of 2020: Visions of En-

gineering in the New Century (National Academy,

Washington 2004)

44.18 D. Oblinger: Boomers, Gen-Xers, and Millennials:

Understanding the “New Students”, EDUCAUSE Rev.

38, 36–45 (2003)

44.19 B.S. Heck, D.S. Dorato, D.S. Bernstein, C.C. Bissell,

N.H. McClamroch, J. Fishstrom: Future directions in

control education, IEEE Control Syst. Mag. 19,36–53

(1999)

44.20

G. Clough: Educating the Engineer of 2020: Adapt-

ing Engineering Education to the New Century

(National Academy, Washington 2005)

44.21 T.E. Duncan, B. Pasik-Duncan: Stochastic adap-

tive control. In: The Control Handbook,ed.by

W.S. Levine (CRC, Boca Raton 1995) pp. 1127–1136

44.22 T.E. Duncan, B. Pasik-Duncan: Adaptive control

of continuous time stochastic systems, J. Adapt.

Control Signal Process. 16, 327–340 (2002)

44.23 B. Pasik-Duncan: Stochastic systems. In: Wiley En-

cyclopedia of IEEE, Vol. 20, ed. by J.G. Webster

(Wiley, New York 1999) pp. 543–555

44.24 F. Black, M. Scholes: The pricing of options and

corporate liabilities, J. Polit. Econ. 81,637–654

(1973)

44.25 R.C. Merton: Theory of rational option pricing, Bell

J. Econ. Manag. Sci. 4, 141–183 (1973)

44.26 S. Crepin, D. Houinato, B. Nawana, G.D. Avode,

P. Preux, J. Desport: Link between epilepsy and

malnutrition in a rural area of Benin, Epilepsia 48,

1926–1933 (2007)

44.27 T.E. Djaferis: Automatic control in ﬁrst-year en-

gineering study, IEEE Control Syst. Mag. 24,35–37

(2004)

44.28 D.S. Bernstein: Innovations in undergraduate ed-

ucation, Special Issue, IEEE Control Syst. Mag. 24,

1–101 (2004)

44.29 D.S. Bernstein: Innovations in undergraduate edu-

cation: Part II, Special Issue, IEEE Control Syst. Mag.

25, 1–106 (2005)

44.30 K.J. Åström: Challenges in control education, 7th

IFAC Symposium on Advances in Control Education

(Universidad Politecnica de Madrid, 2006)

44.31 DARPA: http://www.darpa.mil/grandchallenge/ Vol.

2008: Defense Advanced Research Projects Agency

(DARPA, 2007)

44.32 The MathWorks Inc.: Using MATLAB. The Language

of Technical Computing (The MathWorks Inc., Natick

2002)

44.33 K.J. Åström, R.E. Klein, A. Lennartson: Bicycle dy-

namics and control, IEEE Control Syst. Mag. 25,

26–47 (2005)

44.34 M.W. Spong: Project based control education, 7th

IFAC Symposium on Advances in Control Education

(Universidad Politecnica de Madrid, 2006)

44.35 P.J. Gawthrop, E. McGookin: Using LEGO in con-

trol education, 7th IFAC Symposium on Advances

in Control Education (Universidad Politecnica de

Madrid, 2006)

44.36 J. LaCombe, C. Rogers, E. Wang: Using Lego Bricks

To Conduct Engineering Experiments (American So-

ciety for Engineering Education, Salt Lake City 2004)

44.37 N. Jaksic, D. Spencer: An Introduction To Mecha-

tronics Experiment: Lego Mindstorms Next Urban

Challenge (American Society for Engineering Edu-

cation, Honolulu 2007)

44.38 D. Hansen, B. Self, B. Self, J. Wood: Teaching Un-

dergraduate Kinetics Using A Lego Mindstorms Race

Part E 44

778 Part E Automation Management

Car Competition (American Society for Engineering

Education, Salt Lake City, 2004)

44.39 B.S. Heck, N.S. Clements, A.A. Ferri: A LEGO exper-

iment for embedded control system design, IEEE

Control Syst. Mag. 24,43–56(2004)

44.40 D.S. Bernstein: The quanser DC motor control

trainer – individual or team learning for hands-

on education, IEEE Control Syst. Mag. 25,90–93

(2005)

44.41 S. Dormido: The role of interactivity in control

learning, 6th IFAC Symposium on Advances in Con-

trol Education (Oulu, 2003)

44.42 J.L. Guzman, K.J. Åström, S. Dormido, T. Hagglund,

Y. Piguet: Interactive learning modules for PID

control, 7th IFAC Symposium on Advances in Con-

trol Education (Universidad Politecnica de Madrid,

2006)

44.43 M. Johansson, M. Galvert, K.J. Åström: Interac-

tive tools for education in automatic control, IEEE

Control Syst. Mag. 18, 33–40 (1998)

44.44 B. Pasik-Duncan: Workshop for Czech high school

students and teachers at the 16th IFAC world

congress, IEEE Control Syst. Mag. 26, 110–111 (2006)

44.45 M.H. Shor, F.B. Hanson: Bringing control to stu-

dents and teachers, IEEE Control Syst. Mag. 24,

20–30 (2004)

44.46 B. Pasik-Duncan: Mathematics education of to-

morrow, Assoc. Women Math. (AWM) Newslett. 34,

6–11 (2004)

44.47 D. Duncan, B. Pasik-Duncan: Undergraduates’

partnership with K-12, American Control Confer-

ence (Anchorage, 2002) pp. 1103–1107

44.48 M.K. Michael: Encouraging young women toward

engineering and applied sciences: ISIC/MED special

preconference workshop, IEEE Control Syst. Mag.

26, 100–101 (2006)

44.49 B. Pasik-Duncan, T. Duncan: KU stochastical adap-

tive control undergraduate success stories, Am.

Control Conf. (ACC) (Anchorage 2002) pp. 1085–

1086

44.50 IFAC Technical Committee on Control Education (ED-

COM):

http://www.grifﬁth.edu.au/centre/icsl/edcom/

(last accessed February 19, 2009)

44.51 D. Purkerson Hammer, S.M. Paulsen: Strategies and

processes to design an integrated, longitudinal

professional skills development course sequence,

Am. J. Pharmaceut. Educ. 65, 77–85 (2001)

44.52 B. Pasik-Duncan, T.E. Duncan: Research experience

at all levels, American Control Conference (Maui,

2003) pp. 3036–3038

44.53 Illinois State University: http://www.sotl.ilstu.edu/

(last accessed February 19, 2009)

44.54 K. Lewin: Action research and minority problems,

J. Social Issues 2, 34–46 (1946)

44.55 E.L. Boyer: Scholarship Reconsidered: Priori-

ties of the Professoriate (Princeton Univ. Press,

Lawrenceville 1990)

Part E 44

779

Software Man

45. Software Management

Peter C. Patton, Bijay K. Jayaswal

This chapter is an introduction to software

management in the context of automation. It

recognizes how software and automation are in-

tertwined and have been mutually enabling in

enhancing the reach and seemingly unimaginable

applications of these two disciplines. It further

identiﬁes software engineering as application of

various tools, techniques, methodologies, and dis-

ciplines to produce and maintain an automated

solution to a problem and how software manage-

ment plays a central role in making it possible. We

recognize that software must be managed like any

other corporate or organizational resource albeit as

a virtual rather than an actual or tangible entity.

In this chapter we restrict ourselves to three cru-

cial issues of software management in the context

of software as a component in automation and

how it enhances its value and availability by ef-

fective software distribution, asset management,

and cost estimation. It presents current best prac-

tices in software automation, distribution, asset

management, and cost estimation.

45.1 Automation and Software Management . 779

45.1.1 Software Engineering

and Software Management ......... 780

45.2 Software Distribution............................ 781

45.2.1 Overview of Software

Distribution/Software Delivery ..... 781

45.2.2 Software Distribution

in MS Conﬁguration Manager 2007 782

45.2.3 On-Demand Software ................. 785

45.2.4 Electronic Software Delivery......... 786

45.3 Asset Management ............................... 786

45.3.1 Software Asset Management

and Optimization ....................... 786

45.3.2 The Applications Inventory

or Asset Portfolio........................ 787

45.3.3 Software Asset Management Tools 787

45.3.4 Managing Corporate Laptops

and Their Software ..................... 788

45.3.5 Licence Compliance Issues

and Beneﬁts.............................. 788

45.3.6 Emerging Trends

and Future Challenges ................ 789

45.4 Cost Estimation .................................... 789

45.4.1 Estimating Project Scope ............. 789

45.4.2 Cost Estimating for Large Projects . 789

45.4.3 Requirements-Based

aprioriEstimates....................... 791

45.4.4 Training Developers

to Make Good Estimates.............. 791

45.4.5 Software Tools for Software

Development Estimates............... 792

45.4.6 Emerging Trends

and Future Challenges ................ 793

45.5 Further Reading ................................... 794

References .................................................. 794

45.1 Automation and Software Management

Automation and software are remarkably intertwined.

As organized disciplines, they have independent ori-

gins but are deeply interlocked now. Software, as an

integral component of computers, is an essential tool

for automation of a wide range of industrial, enter-

prise, educational, scientiﬁc, military, medical, home,

entertainment, and personal devices, processes, and

equipment. On the other hand, automation software,

compilers, and a range of computer-assisted software

engineering (CASE) tools are crucialfor estimating, de-

veloping, testing, maintaining, and integrating all kinds

of software, especially large and complex ones. CASE

Part E 45

780 Part E Automation Management

Solution Coding and testing

Solution

inception

Software

elaboration

Software

solution

Software

transition

Time

Objectives

milestone

Architecture

milestone

Software

capability

milestone

Software

product

release

Solution

needs and

implied

requirements

Design and

analysis

Reconfiguration

and change

management

Fig. 45.1 Phases and milestones of software development

automation tools are nothing but computer programs

that assist software analysts, engineers, and coders dur-

ing all phases of the software and system development

lifecycle(Fig.45.1). They are used to automatesoftware

development process and to ease the task of coordi-

nating various events in the development cycle. CASE

tools are usually divided into two main groups: those

that deal with the upstream parts of the system devel-

opment lifecycle (preliminary estimating, investigation,

analysis, and design) are referred to as front-end CASE

tools, and those that deal mainly with the implementa-

tion and installation are referred to as back-end CASE

tools [45.1]. Thus software is indispensable for automa-

tion and automation software tools are essential for

developing reliable and cost-effective software.

In its earliest applications, automation was mech-

anization of certain operation or a series of mechanical

tasks and operationsthat would be done manuallyother-

wise. While automation developed initially to meet the

challenges to automate mechanical and manual opera-

tions in industries, software’s origins could be traced

to solving (computing) military and subsequently, sci-

entiﬁc and business-related data-based problems. It is

amazing what a crucial role software has come to play

in computing, automation, and other numerous appli-

cations, given that it was actually an afterthought to

hardware. The term software was not even used un-

til 1958 almost two decades after the invention of the

ENIAC computer in 1940s.

45.1.1 Software Engineering

and Software Management

Software as a tool in automation has enabled a quantum

leap in improving quality, reliability, cost, precision,

safety, and security of a large number of products and

systems and in expanding automation to seemingly

unimaginable applications in the last several decades.

Software has enabled applications such as the Inter-

net and mobile platforms that have transformed our

lives. However, there are huge intellectual, resource,

and research challenges to deliver and control com-

plex systems involving systems of systems, networks

of networks, and agents of agents. They create huge

intellectual control and software management prob-

lems [45.2]. Automation and component technology

would play a critical role in meeting these software

management challenges.

Software consists of computer programs, proce-

dures, and (possibly) associated documentation and

data pertaining to the operation of a computer sys-

tem [45.3]. It is important to note that software is not

just codes or even programs. Further, software runs on

hardware and has been, relatively speaking, a laggard

as regards quality, delivery, and cost relative to hard-

ware part of computers. Software is thus the crucial

element to be addressed in determining cost and per-

formance of a large number of computer applications.

Often computers operate as part of communication

networks, increasingly the web and the Internet, that

provide connectivity for importing and exporting data

for a wide variety of commercial, military, educa-

tional, entertainment, and public services. The limits

of software availability and trustworthiness are simi-

larly determining factors as regards effectiveness and

viability of an increasingly large number of automation

applications.

Software engineering, on the other hand, could

be described as the application of tools, techniques,

methodologies, and disciplines to produce and main-

tain an automated solution to a problem. Designing

and delivering trustworthy software is one of the great

technological challenges of our times. It would require,

among others, the implementation of a lifecycle ap-

proach to software development that addresses software

trustworthiness at upstream phases of the software de-

velopment process [45.3].

That brings us to the subject of software man-

agement. What exactly is software management? An

inclusive view of a software business should include or-

ganizational, strategic, and competitive contexts of an

enterprise processes and technologies that are used in

developing software. That is a vast area and beyond

the scope of this Handbook and indeed this chapter.

Even the software development process is a large disci-

pline consistingof tools,techniques, and methodologies

Part E 45.1

Software Management 45.2 Software Distribution 781

that aid planning, estimating, stafﬁng, organizing, and

controlling the software development process [45.3,4].

In this chapter we would restrict ourselves to three

crucial issues of software management in the con-

text of software as a component in automation and

how it enhances its value and availability by effective

software distribution, asset management,andcost esti-

mation.

45.2 Software Distribution

45.2.1 Overview of Software

Distribution/Software Delivery

Software distribution is a generic term used to describe

automated or semiautomated distribution or delivery of

software, usually on a network including the Internet.

Such facilities enable software providers to distribute,

ﬁx, and update software packages to clients and servers

on an organization’s network. It is usually conﬁgured

to install software with no user intervention and as

such can be used to keep organization’s network up

to date with minimum disruptions. Enterprises face

the challenges of maintaining security and compli-

ance requirements while keeping pace with meeting

regulatory and technological changes. The information

technology (IT) staff has to constantly manage new

security threats and patches, updates as well as innova-

tions. Business success depends on maintaining control

in the face of all these changes. A smart conﬁgura-

tion management suite must meet several requirements

(Fig.45.2) [45.5]:

1. Increase efﬁciency and save time with a complete

hardware and software management solution for all

the systems and users in your complex network en-

vironment.

2. Reduce costs and the demands on help desk re-

sources with tools that help you securely and easily

support users in any networked environment.

3. Protect user productivity and reduce resource needs

by easily keeping up with patches and updates and

maintaining system-level security.

4. Save time and network bandwidth with patented,

ultra-efﬁcient, and fault-tolerant software distribu-

tion technologies.

5. Decrease software licensing costs and quickly

respond to audits with comprehensive software li-

cence monitoring capabilities.

6. Increase efﬁciency and save time by easily mi-

grating users and their proﬁles to new operating

systems.

Software distribution is one of the important tools

of conﬁguration management tools. The rate and scale

of change is increasing. Demands such as Microsoft

Vista migrations, data center consolidation initiatives,

and stringent time-to-market requirements have placed

a new emphasis on the ability to execute changes ef-

fectively and efﬁciently. At the same time, IT is faced

with resource constraints, a geographically dispersed

and mobile workforce, and ongoing security threats.

HP conﬁguration management solutions enable IT

to respond to these demands through automated deploy-

ment and continuous management of software, includ-

ing operating systems, applications, patches, content,

and conﬁguration settings, on the widest breadth and

largest volume of devices throughout the lifecycle for:

•

IT efﬁciency to control management costs

•

Agility to bring services to customers and users

faster.

Software distribution is essentially the process of

making the software products and services available to

users in a manner that meets their cost, quality, and de-

livery expectations. It means that the software provider

successfully delivers the product that the users need,

when they and in the form need it, and at the price

they are willing to pay for it. It also implies that the

OS imaging

and migration

Inventory/IT asset

management

Software

distribution

Management

gateway

Reporting

and dashboard

Software licence

monitoring

Discovery

Remote control/

problem

resolution

Systems

management

Fig. 45.2 A software management console (after [45.5] with per-

mission)

Part E 45.2

782 Part E Automation Management

software provider is able to do so in a cost-effective

manner that is economically and competitively viable.

While for many physical products we take this process

for granted, for software, especially, large and complex

ones, it is quite a challenge. The challenge comes of-

ten from software’s complexity and the very nature of

software design and use.

In addition to what was stated above regarding soft-

ware distribution beneﬁts it also ensures ubiquity and

currency of software updates, and allows for certain

compliance with all licensing provisions, as well as

economy as just licences currently in use will be paid

for.

45.2.2 Software Distribution

in MS Conﬁguration Manager 2007

Microsoft’s System Center Conﬁguration Manager

(MSCCM) is a state-of-the-art toolset for software dis-

tribution and management in a networked organization.

We cite this product extensively as an exemplar of

technology available today without loss of generality,

even though it is a particular proprietary product [45.6].

Other quality software distribution products in this gen-

eral area include LANDesk [45.7], and Akamai [45.8].

Microsoft System Center is a suite of IT management

solutions that help IT departments proactively plan, de-

ploy, manage, and optimize the software lifecycle of

a networked IT environment.

Planning Software Distribution

Planning is the ﬁrst step in effectively upgrading the

network server infrastructure. During this phase, the

IT department must collect critical information about

server infrastructure, including:

•

Assessing the current state of the server infrastruc-

ture and datacenter

•

Identifying each asset that comprises the infrastruc-

ture

•

Identifying the purpose of each asset.

For most organizations, collecting accurate infor-

mation about server assets within the datacenter is

easier said than done. Networked datacenters grow in-

creasingly complex daily as companies introduce and

implement new technology to enhance business perfor-

mance. This makes it difﬁcult for the IT department to

maintain accurate records of server assets, which also

makes it difﬁcult to plan upgrades and enhancements to

server infrastructure. Microsoft System Center (MSCC)

delivers capabilities that make it easier for the IT or-

ganization to collect information needed for in-depth

knowledge of existing infrastructure.

The ﬁrst step of planning a server upgrade is to

identify all the assets that make up the network. The

IT department needs a centralized management solution

that automatically identiﬁes software assets. Microsoft

System Center Conﬁguration Manager 2007 (MSCCM

2007) simpliﬁes this task with hardware and software

inventory capabilities that identify hardware and soft-

ware assets, catalog who is using those assets, and

understand where they are located. Through asset in-

telligence, MSCCM 2007 presents a clear picture of

IT assets by providing identiﬁcation and categorization

of the servers, desktops, laptops, mobile devices, and

software installed across both physical and virtual en-

vironments. Within the datacenter, this provides a fast

method for understanding what server devices are in use

today and who is using them. A new feature available in

the ﬁrst service pack for MSCCM 2007 also enables as-

set intelligence to identify new and changing systems

and notify IT administrators of changes. This can re-

duce time spent identifying and tracking assets during

and after an upgrade project.

As IT organizations move through the phases of

the infrastructure optimization model, planning a server

upgrade presents an opportunity to cut both medium-

and long-term costs by optimizing the use of server re-

sources within the datacenter. Virtualization is one of

the most important trends that can impact server re-

source optimization by changing how an IT department

manages servers and workloads. Virtual machine tech-

nology decouples the physical hardware from software

so that the IT department can run multiple virtual ma-

chines on a single physical server.

Microsoft System Center Operations Manger 2007

and Microsoft System Center Virtual Machine Man-

ager 2007 help the IT department identify how servers

are being used, how each server is performing, and

how each server can be used to its fullest poten-

tial. System Center Operations Manager 2007 monitors

server health and stores vital performance information

in a database that System Center Virtual Machine Man-

ager 2007 can access and analyze. Virtual Machine

Manager 2007 then generates a consolidation report

that provides an easy-to-understand summary of the

long-term performance of a workload. This informa-

tion helps project teams make informed decisions about

which servers would be ideal candidates for consoli-

dation. Also, information about the performance of the

hardware running virtualized applications provides data

that decision-makers need to smartly move those appli-

Part E 45.2

Software Management 45.2 Software Distribution 783

cations off one server onto another, re-image the server,

and then return the applications while maintaining full

availability of the datacenter resources.

Microsoft System Center Data Protection Manager

2007 helps companies plan a server upgrade with conﬁ-

dence by enablingthe IT department to back up existing

data. System Center Data Protection Manager 2007 was

built to protect and recover:

•

Microsoft SQL Server

•

Microsoft Exchange Server

•

Microsoft Ofﬁce SharePoint Server

•

Microsoft Virtual Server

•

Microsoft Active Directory directory service

•

Windows ﬁle services.

Conﬁguring Software Distribution

When the IT department has created an accurate inven-

tory of server assets, it can then design the datacenter

and determine which changes should be made to guar-

antee the most cost-efﬁcient infrastructure. Then further

steps will enable the department to successfully deploy

the Windows Server 2008 operating system, Microsoft

SQL Server 2008, and Exchange Server 2007 SP1 and

transform the datacenter into a strategic asset. During

the conﬁgurationor build phase, the IT departmentmust

create server images, convert physical servers to vir-

tual servers, create a disaster recovery plan, and monitor

the testing process. The build phase offers an opportu-

nity for IT departments to identify areas for reducing

costs, improving efﬁciency, and supporting compliance

efforts. One way to accomplish this is by creating

standardized server images for all server components,

for both physical and virtual machines. System Center

Operations Manager 2007 and System Center Virtual

Machine Manager 2007 facilitate this process.

The task sequencer, driver packages, and dynamic

driver catalog included with Conﬁguration Manager

2007 signiﬁcantly reduce the number of server images

that the IT organization must create and be deployed

to either physical or virtual machines. IT administrators

can create a simple generic image and dynamically add

the necessary drivers during the build. In addition, by

integrating vendor-provided tools, Conﬁguration Man-

ager 2007 can automate the setup of redundant array

of independent disks (RAID), storage area network

(SAN), and Internet small computer system interface

(iSCSI) hard-drive conﬁgurations as part of the task se-

quence. This can have a favorable impact on the amount

of work required later as upgrades are issued. Upon

creation of the server images for physical machines,

Virtual Machine Manager 2007 converts the appropri-

ate images for virtual machines. Traditionally, this task

can be slow and disrupt business operations, but Virtual

Machine Manager 2007 uses the volume shadow copy

service, which helps administrators create virtual ma-

chines without interrupting the source physical server.

Virtual Machine Manager 2007 also simpliﬁes this

whole process by providing a task-based wizard that

helps guide administrators. Once images are created,

Virtual Machine Manager 2007 supports a complete li-

brary that organizes and manages all the building blocks

of the virtual datacenter within a single interface.

Data Protection Manager 2007 helps prevent the

IT department from losing critical business data when

upgrading server infrastructure. By integrating a point-

in-time database restore with existing application logs,

Data Protection Manager can deliver nearly zero data

loss recovery for Microsoft Exchange Server, SQL

Server, and SharePoint Server, eliminating the need to

replicate or synchronize data. Data Protection Manager

also uses both disk and tape media to enable fast re-

store from disk and supports long-term data retention

and off-site portability with disks.

Before deploying upgrades to the server environ-

ment, the IT department must perform tests to ensure

business continuity when the new server products go

live. System Center Operations Manager 2007 makes

it easy to access the results of these tests, much in

the same way that it monitors the overall health of the

server infrastructure. An IT department can also cre-

ate scenarios that act like the end-user of a speciﬁc

service to monitor success and failure rates and perfor-

mance statistics results that can help identify potential

deployment issues. In addition, administrator-simulated

end-users can access Virtual Machine Manager by way

of a web portal that is designed for user self-service.

This portal enables test users and development users to

quickly provision new virtual machines for themselves,

according to the controls set by the administrator. Not

only can IT personnel quickly test new conﬁgurations,

but they can also uncover problems before deployment.

Managing Software Distribution

During deployment, IT departments must quickly roll

out new products while remaining agile so they can re-

spond to changes. Costs must also be kept to a minimum

and business operations must not be disrupted. In the

past, deploying new server software required someone

to sit down at each server and complete the upgrade.

This manual process took signiﬁcant resources and did

not guarantee that servers were deployed with con-

Part E 45.2

784 Part E Automation Management

sistent conﬁgurations. Determining which virtual and

physical machines to link together was also difﬁcult

because companies did not have the data, such as work-

loads, performance metrics, and network capacity, to

create optimal arrangements. Companies often risked

losing vital company data during the migration process.

System Center helps alleviate these challenges.

With Conﬁguration Manager 2007, IT administra-

tors can roll out new servers rapidly and consistently

by automating operating system deployments and task

sequences. IT administrators can fully deploy and con-

ﬁgure servers from previous states, either by updating

or replacing original equipment manufacturer (OEM)

builds, or by installing the operating system and appli-

cations on new computers. Preboot execution environ-

ment protocol and Windows deployment services also

make it easier to deploy servers that have no operating

system installed: just plug in the server and turn it on.

The task sequencer in Conﬁguration Manager 2007

fully automates the end-to-end deployment process, en-

abling zero-touch to near-zero-touch deployments. This

means that the process ofbuildingservers, whichcan in-

clude more than 80 steps, including image loads, driver

loads, update loads, and multiple reboots, can be han-

dled by Conﬁguration Manager automatically.

IT departments can also maintain visibility of the

state of the infrastructure throughout the entire datacen-

ter deploymentand management process.Conﬁguration

Manager 2007 generates detailed reports about the de-

ployments and provides information about those that

have failed. This information helps the IT department

resolve problems quickly, easily, and proactively. To

maximize server utilization, it is critical that IT admin-

istrators select the appropriate virtual machine host for

a given workload. Virtual Machine Manager 2007 helps

IT departments with this complex task of intelligent

placement. Virtual Machine Manager 2007 uses a holis-

tic approach to selecting the appropriate hosts based on

four factors:

1. The resource consumption characteristics of the

workload

2. Minimum central processing unit (CPU), disk,

random-access memory (RAM), and network ca-

pacity requirements

3. Performance data from virtual machine hosts

4. Preselected business rules and models associated

with each workﬂow that contain knowledge from

the entire lifecycle of the workload.

After the analysis, Virtual Machine Manager 2007

produces an intelligent placement report that helps the

IT department select the appropriate host for a given

workload.

As IT administrators migrate information to an up-

dated server platform, it is critical that data is not lost

or corrupted. Once the new platform is in place, Data

Protection Manager 2007 will identify the new server

environment and enable customers to quickly and eas-

ily restore the data where it needs to go. Administrative

delays associated with restores are also reduced by us-

ing a restore user interface that is based on the calendar,

robust media management functionality, and disk-based

end-user recovery. With Data Protection Manager 2007,

restoring informationtakes seconds and involves simply

browsing a share and copying directly from Data Pro-

tection Manager to the production server. By enabling

customers to restore data from disk, Data Protection

Manager signiﬁcantly shortens the amount of time it

takes to recover data, allowing customers to recover

data in minutes versus the hours it takes to recover from

tape. Data Protection Manager also minimizes the risk

of failure that is associated with recovering data from

tape.

Monitoring Software Distribution

After successfully upgrading the server infrastructure

with next-generation server technology from Microsoft,

the IT department must continue to monitor the infras-

tructure to ensure technology and licences are up to

date, the network is secure, and commitments to meet

service level agreements for performance and availabil-

ity are met. In addition, the IT department must ensure

consistency within server conﬁgurations, for example,

guaranteeing that every exchange server has the same

conﬁguration and that server resources are being used

with maximum efﬁciency to derive the most value from

existingresources. Meeting these goals was once a chal-

lenge because the IT department did not have a solution

that enabled the management of the entire server infras-

tructure from a central location.

System Center Server Management Suite Enterprise

not only simpliﬁes and speeds the deployment of new

server software, it also eases theongoing task of manag-

ing the entireserver infrastructure ona day-to-daybasis.

Centralized Management of Server Networks System

Center offers many ways for IT departments to proac-

tively manage the state of IT infrastructure regardless of

its complexity; for example, System Center Operations

Manager 2007 provides an easy-to-use management en-

vironment that can oversee thousands of servers and

applications, delivering a comprehensive view of the

health of the datacenter. System Center Operations

Part E 45.2