CHAPTER 1: Network Fundamentals 6 massive changes in the later part of the twentieth century. By looking at these changes, you will see the development of OSes, hardware, and innova- tions that are still used today. Early Telecommunications and Computers Telecommunications got its start in 1870s in Brantford Ontario, when Alexander Graham Bell developed the idea of a telephone. After the first successful words were sent over the device on March 10, 1876, a revolu- tion of communication began. Within decades of its conception, millions of telephones were sold, with operators connecting people using manual circuit switching. This method of calling the operator to have them connect you to another party was routine until the mid-twentieth century, when mechanical and electronic circuit switching became commonplace. These events would have a massive impact on the innovation of computers, even though they wouldn’t be invented until 60 years after Bell’s first successful phone call. Although arguments could be made as to whether ancient devices (such as the abacus) could be considered a type of computer, the first computer that could be programmed was developed by a German engineer named Konrad Zuse. In 1936, Zuse created the Z1, a mechanical calculator that was the NOTES FROM THE FIELD… Knowledgeable Network Users The Internet is a vast network of interconnected com- puters that your computer becomes a part of when- ever it goes online. Because more people than ever before use the Internet, this means that many people are familiar with the basic concepts and features of networking without even being aware of it. This is a particular benefit in training the users of a network, as many will be familiar with using e-mail, having user accounts and passwords, and other technologies or procedures. Unfortunately, a little knowledge can also be a dangerous thing. When dealing with knowledgeable users, it is important to realize that they may have developed bad habits. After all, a user with years of experience will have found a few shortcuts and may have gotten lazy in terms of security. For example, the user may use easy- to-remember passwords that are easy to guess. In such a case, the solution would be to implement policies on using strong passwords (passwords with at least eight characters consisting of numbers, upper and lowercase letters, and non-alphanumeric characters), changing passwords regularly, and not sharing passwords with others. If your company has an intranet, you can pro- vide information on such policies to employees. Another problem is that users may attempt to per- form actions that aren’t permitted in an organization, such as installing unapproved software or accessing restricted data. It is also important to setup security on a network so users can only access what they need to perform their job. This minimizes the chance that someone might modify or delete a file, view sensitive materials that are meant to be confidential, or install software that contains malicious programming or isn’t work related (such as games). Even in an environment with a trusting atmosphere, accidents happen and problems arise. Setting up proper security can prevent avoidable incidents from occuring. What Is a Network? 7 first binary computer. Zuse continued making innovations to his design, and five years later had reached the point where the Z3 was able to accept programming. Although the next version of his computer would use punch cards to store programs, Zuse used movie film to store programming and data on the Z3 due to a supply shortage of paper during World War II. Just as his computers evolved, so did his programming skills. Zuse’s achievements also extended to creating the first algorithmic programming language called Plankalkül, which later was used to create the first computer chess game. During this same time, John Atanasoff and Clifford Berry developed what is acknowledged to be the first electronic-binary computer. Created at the University of Iowa, the initial prototype acquired this team a grant that allowed them to build their 700 pound final product, containing more than 300 vacuum tubes and approximately one mile of wire. Because the war prevented them from completing a patent on their computer, the computer was dismantled when the physics department needed storage space that was being used by the machine. The distinction of being first initially went to John Mauchly and J. Presper Eckert for their Electrical Numerical Integrator And Calculator (ENIAC I) computer, until a 1973 patent infringement case determined Atanasoff and Berry were the first. The ENIAC I was developed with funding from the U.S. government and based on the work of John Atanasoff. Starting work in 1943, the project took two and a half years to design and build ENIAC I, at a cost of half a million dollars. The ENIAC I was faster than previous computers, and used to per- form calculations for designing a hydrogen bomb, wind-tunnel designs, and a variety of scientific studies. It was used until 1955 when the 30-ton, 1,800 square foot computer was ultimately retired. Another computer that was developed during this time was the MARK I computer, developed by Howard Aiken and Grace Murray Hopper in 1944 in a project cosponsored by Harvard University and International Business Machines (IBM). Dwarfing the ENIAC at a length of 55 feet and five tons in weight, the MARK I was the first computer to perform long calculations. Although it was retired in 1959, it made a lasting mark on the English lan- guage. When the MARK I experienced a computer failure, Grace Murray Hopper checked inside the machine and found a moth. She taped it to her log book and wrote “first actual bug found”, giving us the terms “bug” for a computer problem, and “debug” for fixing it. In 1949, Hopper went on from the MARK I project to join a company created by John Mauchly and J. Presper Eckert, which was developing a 1,500 square foot, 40-ton computer named UNIVersal Automatic Computer (UNIVAC). UNIVAC was the first computer to use magnetic tape instead of paper cards to store programming code and data, and much faster than CHAPTER 1: Network Fundamentals 8 the previous computers we’ve discussed. Although the MARK I took a few seconds to complete a multiplication operation and ENIAC I could perform hundreds of operations per second, UNIVAC could perform multiplication in microseconds. What made UNIVAC popular in the public eye, how- ever, was a 1952 publicity stunt where the computer accurately predicted the outcome of the presidential election. Although Dwight Eisenhower and Adlai Stevenson were believed evenly matched going into the November 4 election night, UNIVAC predicted that Eisenhower would get 438 electoral votes, while Stevenson would only get 93. In actuality, Eisenhower got 442 electoral votes, while Stevenson got 89. Although political analysts had been unable to predict the outcome, UNIVAC did so with a one percent margin of error. Having earned its place in history, the original UNIVAC currently resides in the Smithsonian Institute. Although UNIVAC was the more successful computer of its day, 1953 saw IBM release the EDPM 701. Using punch cards for programs, 19 of these were sold (as opposed to 46 UNIVACs sold to business and govern- ment agencies). However, development of this computer lead to the IBM 704, considered to be the first super-computer. Because it used magnetic core memory, it was faster than its predecessor. This series of computers further evolved to the development of the 7090 computer in 1960, which was the first commercially available computer to use transistors, and the fastest computer of its time. Such innovations firmly placed IBM as a leader in computer technology. The Space Age to the Information Age Although IBM and the owners of UNIVAC were contending for clients to buy their computers, or at least rent computer time, the former USSR launched Sputnik in 1957. Sputnik was the first man-made satellite to be put into orbit, which started a competition in space between the USSR and the United States, and launched a number of events that made advances in computers. Although author Arthur C. Clark had published an article in 1945 describing man-made satellites in geosynchronous orbit being used to relay transmis- sions, communication satellites didn’t appear until after Sputnik’s historic orbit. In 1960, Bell Telephone Laboratories (AT&T) filed with the FCC to obtain permission to launch a communications satellite, and over the next five years, several communication satellites were orbiting overhead. Obviously, the most notable result of Sputnik was the space race between the U.S. and USSR, with the ultimate goal of reaching the moon. The U.S. started the National Aeronautics and Space Administration (NASA), began launching space missions, and achieved the first manned landing on the moon in 1969. Using computers that employed only a few thousand lines of What Is a Network? 9 code (as opposed to the 45 million lines of code used in Windows XP), the onboard computer systems provided necessary functions and communicated with other computers on earth. Communications between astronauts and mission control on earth also marked the furthest distance of people com- municating to date. The cold war and space race also resulted in another important milestone in computer systems and communication systems. As we discussed earlier, the U.S. government started the Advanced Research Projects Agency, which developed such important technologies as follows: ARPANet, the predecessor of the modern Internet, which connected multiple institutions and areas of government together for research purposes. Packet Switched Networks, where messages sent over a network are broken into packets. They are then sent over the network, and reas- sembled after reaching the destination computer. TCP/IP, which specifies rules on how data is to be sent and received over the network, and provides utilities for working over a network. Although only a few educational institutions and the government were networked together through ARPANet, this led to the first e-mail program being developed in 1972, and the first news server being developed in 1979. The Internet was years away, but its foundation was set here. Hardware and Operating Systems Major advances pushed the development of computers and networking in the 1970s. In 1971, Intel produced the first microprocessor, which had its own arithmetic logic unit and provided a way of creating smaller, faster com- puters. It was the first of many Intel processors produced over the years, including the: 8008 processor produced in 1972. 8080 processor (an 8-bit processor) produced in 1974. 8086 processor (a 16-bit processor) produced in 1978. Because other technology needs to catch up to the speed of the processor, an 8088 processor (an 8-/16-bit processor) is released in 1979. It isn’t until 1983 that IBM releases the XT with the 8086 processor (and option to add an 8087 math co-processor). 80286 (16-bit processor) produced in 1982. 80386 (32-bit processor) produced in 1985. CHAPTER 1: Network Fundamentals 10 80486 (32-bit processor) produced in 1989. Pentium (32-bit processor) produced in 1993, and which ended the x86 naming scheme for their processors. After this, Intel chips bore the Pentium name, inclusive to the Pentium 75 (in 1994), Pentium 120, 133, and Pentium Pro 200 (in 1995), Pentium MMX and Pentium II (in 1997), and Pentium III (in 1999). As you would expect, each generation was faster than the last. Just as processing changed in the 1970s, so did storage. In 1973, IBM developed the first hard disk, and an 8” floppy drive, replacing the need to store data and programs solely on magnetic tapes. This massive floppy was quickly replaced by the 5.25” floppy in 1976, which was later succeeded by the 3.5” floppy disk that was developed by Sony in 1980. These methods of storing data became commonplace until 1989 when the first CD-ROM was developed, and again changed in 1997 with the introduction of DVDs. With the advances in technology, it was only a matter of time before someone developed a home computer. Prior to the mid-1970s, computers were still too large and expensive to be used by anyone but large corporations and governments. With the invention of the microprocessor, a company called Micro Instrumentation and Telemetry Systems (MITS) developed the Altair 8800 using the Intel 8080 processor. Although it included an 8” floppy drive, it didn’t have a keyboard, monitor, or other peripherals that we’re accustomed to today. Programs and data entry were entered using toggle switches at the front of the machine. Although it couldn’t be compared to personal computers of today, it did attain the distinction of being the first. The Altair also provides a point of origin for Microsoft, as Bill Gates and Paul Allen developed a version of the BASIC programming language for Altair that was based on a public domain version created in 1964. Microsoft went on to create an OS that required users to type in commands called PC-DOS for the first IBM computer named The Acorn in 1981, but main- tained ownership of the software. This allowed them to market their OS to other computer manufacturers and build their software empire. Microsoft went on to develop OSes such as MS-DOS, Windows 1.0–3.0, Windows for Workgroups, Windows NT, Windows 95, Windows 98, Windows ME, Windows 2000, Windows Server 2003, Windows XP, Windows Server 2008, and Windows Vista. Newly released as of this writing is Windows 7, Microsoft’s newest desktop OS. UNIX was another OS that originated in the 1970s and also led to a case of one OS leading to another. Ken Thomson and Dennis Ritchie of Bell Labs developed UNIX in 1970, which came to be used on high-end servers What Is a Network? 11 and (years later) Web servers for the Internet. Linus Torvalds used UNIX as the basis for developing Linux 20 years later. Linux is open source, mean- ing that the code is available to use for free, and is the only OS that acts as competition for Microsoft and Apple today. Apple Computer’s start also originates in the 1970s, when Steve Jobs and Steve Wozniak incorporated their company on April Fools Day of 1976, and introduced the Apple II the next year. It wasn’t until the 1980s however when Apple really made its mark. In 1981, Xerox developed a graphical user interface (GUI) that used the windows, icons, menus, and mouse support that we’re familiar having in OSes today. However, Xerox never released it to the public. Developing its Apple Lisa and Macintosh on the work done by Xerox, Apple introduced Lisa as the first GUI personal computer in 1983. The Macintosh was found easy to use by the public and made Apple the major competition of IBM after its release. The Windows, Icons, Menus, Pointer (WIMP) interface that gave Apple and Windows their looks wasn’t the only major contribution Xerox made to computers and networking. While working at Xerox’s Palo Alto research center, Bob Metcalfe was asked to develop a method of networking their computers. What he created was called Ethernet. Ethernet was different from other networks like the Internet, which connected remote computers together using modems that dialed into one another, or dumb terminals that had no processing power and were only used to access mainframe computers. It connected computers together using cabling and network adapters, allowing them to communicate with one another over these physical connections. If Ethernet sounds like many networks in use today, you’d be correct; Ethernet is an industry standard. After Ethernet was developed, OSes that were specifically designed for networking weren’t far behind. In 1979, Novell Data Systems was founded with a focus on developing computer hardware and OSes. In 1983, however, Novell changed focus and developed NetWare, becoming an industry leader in network OSes. Unlike other OSes that resided on a computer and could be used as either a standalone machine or a network workstation, NetWare has two components. The NetWare OS is a full OS and resides on a server, which processes requests from a network user’s client machine. The com- puter that the network user is working on can run any number of different OSes (such as Windows 9x, NT, etc), but has client software installed on it that connects to the NetWare server. When a request is made to access a file or print to a NetWare server, the client software redirects the request to the server. Because of its popularity as a network OS, it was widely used on corporate and government networks. CHAPTER 1: Network Fundamentals 12 The Information Age Appears In the 1980s, computers became more commonplace in homes and businesses. Prices had dropped to the point that it was now affordable to have a computer in the home, and powerful enough to be worth having a 286 or 386 computer. Although many people found computers useful, they quickly outgrew the desire to have a standalone machine and wanted to be networked to others. The 1980s and 1990s saw growing popularity in Bulletin Board Sys- tems (BBSs), where one computer could use a modem and telephone line to directly dial another computer. Computers with BBS software provided the ability for users to enjoy many of the features associated with the Internet, including message boards, e-mail, chat programs (to send messages instantly to other users), the ability to download programs and other files, play online games, or other features. Because it was done over a modem, BBSs were largely comprised of other computer users within their community, although message networks were used to have discussions with people in other cities or countries. Although BBSs were eventually replaced by the Internet, the 1980s also saw changes that would affect the future of cyberspace. In 1983, the University of Wisconsin developed the Domain Name System (DNS). DNS provided a way of translating IP addresses used to uniquely identify comput- ers on TCP/IP networks. Using DNS, a number like 207.46.250.222 can be translated to a friendly domain name such as microsoft.com. In 1984, DNS became part of ARPANet, and would eventually play a major part resolving domain names on the Internet. Also during the mid-1980s, the backbone of the Internet was developed. The backbone was a central network that connected other networks on the Internet. Between 1985 and 1988, T1 lines were developed to accommodate a necessary increase of dataflow and speed on the Internet. The speed that data could be sent over the backbone was now increased to 1.544 Mbps. Perhaps the greatest defining moment in the shape of the Internet was Tim Berners-Lee’s creation of HyperText Markup Language (HTML). In 1989, a textual version was created that supported hyperlinks, but this evolved into the language that was released by CERN in 1991, and used to create documents that can be viewed online (i.e. Web pages). In 1993, Mosaic became the first Internet browser, allowing users to view such Web pages, but others such as Netscape and Internet Explorer soon appeared. When ARPANet was retired in 1990, the first company to provide dial-up access to the Internet was formed. The World (www.theworld.com) provided the ability (for a fee) to dial into their system using a modem, and connect to the Internet. In 1993, other Internet service providers (ISPs) appeared that What Is a Network? 13 provided this service, increasing the number of people using the Internet steadily. Although initially used as a repository of programs, research, and other resources, this opened the Internet up to commercial use, which evolved it into the entity we know today. Modern Networking Technologies Since the year 2000, when Y2K was the biggest rage and the e-mail virus Melissa was first introduced, there have been countless developments in net- working. Although there have been many new technologies developed, one thing remains the same – the fundamentals of networking have not changed. We still use models to explain, design, administer, and troubleshoot net- working today and because of those standards more and more proprietary development (closed source) software, systems, and services are starting to disappear. Open source technologies are starting to take hold in the market as support for them grows within the communities that develop them. Some newer technologies that we will cover in this book are multiprotocol label HEAD OF THE CLASS… Virtualization The term virtualization is very broad. In its simplest definition, it’s the term used to explain how hardware resources and OSes are managed using virtualiza- tion software such as VMware ESX Server, Windows Hyper-V, and Citrix Xen to name a few of the most com- monly used systems. There are also many types of virtualization – desktop virtualization, storage virtualization, application virtual- ization, and server virtualization … the list goes on. The underlying theory for all is that the OS is abstracted from the hardware portion of the computer. Now, with a platform between the resources and the OS managing them (VMware, Microsoft Hyper-V, Citrix Xen, etc) the hardware resources can be given to each OS installed and if configured correctly, gives you a foolproof plan for getting systems back up and running quickly. Likely, your virtualized environment will be con- figured to run on a high-speed storage area network (SAN) that can be configured as a high-speed back- bone between your systems, disks, and data. With a RAID array, disk failure, is imminent but solvable. As almost all disks come with a Mean Time Between Fail- ure (MTBF) that renders the disk likely to fail after a certain amount of time, usage, or abuse, it is important to have a solution in place to fix your problems as they occur. With this solution is added with backed up data, you have a bulletproof solution that works at very high speed and is completely redundant. You will, however, pay a price in hardware and software costs as well as deployment time, migration time, and so on. With your OSes configured as simple images (or ISO files), you can also recover from OS malfunctions quickly. Because your data is backed up, you can quickly recreate your systems from “snapshots” of you taken from virtualization software such as VMware as an example. Restoring a system, its data and add- ing more storage, hardware resources, and services becomes very easy to do. As you can see, there are many benefits to learning about and working within virtualized environments. CHAPTER 1: Network Fundamentals 14 switching (MPLS), virtualization, and cloud computing among many others. They are paving the way for even newer breakthroughs to take place over the next year. Some jokingly say that the LAN is dead – could it be true? Mobile networking is also becoming increasingly important because of this. As more and more games, videos, and songs are distributed over mobile devices, the demand to supply faster, more robust ways to support them over the network will take place. As we approach 2010, it’s interesting to look back on the last decade of networking … how it has evolved and where we will wind up next. LOGICAL NETWORKING TOPOLOGIES Because networks vary from one another depending upon a range of factors, it should come as no surprise that there are different network models that can be chosen. The network model you choose will affect a network infra- structure’s design, and how it is administered. Depending on the model or models used, it can have an impact on the location of computers, how users access resources, and the number of computers and types of OSes required. Some of the models and topologies available to choose from are as follows: Centralized Decentralized (Distributed) Peer-to-Peer Client/Server Virtual Private Network (VPN) Virtual Local Area Network (VLAN) Because it’s arguable that there is little need for a network without the use or sharing of resources on it, then we would have to say that all resources would either have to be centralized, decentralized, or multiple networks con- figured and accessible to facilitate both models simultaneously. Note We cover new technologies for completeness, although not all new (or bleeding edge) technology – for example, we cover virtualization topics although you will not be tested on any vendor’s offerings, or how they are configured, deployed, or utilized. As a working Network+ technician, you are sure to see these technologies deployed while you work so you should be aware of them. Logical Networking Topologies 15 Network modeling is important for getting the first leg of network design complete. Will you be centralizing your resources, or will they be decentralized? Centralized When a centralized network model is used, a network’s resources are centrally located and administered. This approach allows network administrators to have better access to equipment and can provide better control over security issues. However, because responsibility for managing these resources now rests with the network administrator or Information Technology (IT) staff, the administrative workload increases. A centralized model will affect the physical location of servers and certain other resources on your network by situating them within a specific area. You’ll remember that servers are computers that accept requests from client computers (which users of the network work on), and provide services and resources that the client has proper access to use. As we’ll discuss later in this chapter, dedicated servers generally have larger hard disks, more memory, and faster processors than the workstations accessing them. When a centralized model is used, these servers are generally located in a secure, central location, such as a dedicated server room. This secured room can also be used to house other resources, such as routers, switches, firewall, Web servers, plotters, and other devices. Because they are stored in a central location, additional work may be required to manage them. For example, let’s say you had a plotter that was kept in a server room. Anytime anyone needed the plotter installed as printer on their computer, you would need to set up permissions for them to use it. If the user sent a print job to this plotter, someone from the IT staff would need to enter the secure room to get their printout. In addition, there would also be the need to replace paper and toners used in the device. In a central- ized model, administration of the resources is also centralized. Despite the previous scenario, in some ways managing resources can be easier with this model. By keeping these resources in one area, a network administrator can easily change backup tapes, replace hard disks, or fix other issues as required. Imagine the issues of having servers in offices throughout a city or region, and having to visit each of them whenever a tape needed to be replaced after running a tape backup. By keeping resources centralized, less work is needed for administration of them. Depending on the requirements of an organization, the centralized net- work model can also mean that fewer servers or other devices are needed. Rather than each building having their own server on the premises, users . It isn’t until 19 83 that IBM releases the XT with the 8086 processor (and option to add an 8087 math co-processor). 80286 (16-bit processor) produced in 1982. 8 038 6 (32 -bit processor) produced. produced in 1985. CHAPTER 1: Network Fundamentals 10 80486 (32 -bit processor) produced in 1989. Pentium (32 -bit processor) produced in 19 93, and which ended the x86 naming scheme for their processors November 4 election night, UNIVAC predicted that Eisenhower would get 438 electoral votes, while Stevenson would only get 93. In actuality, Eisenhower got 442 electoral votes, while Stevenson