Back to Basics: Understanding Operating Systems and Network Operating Systems
Jeff Kabachinski · Biomedical Instrumentation & Technology · 2010
The traditional boundary lines between computer operating systems and network operating systems have blurred. Most of today's operating systems (OS), like Microsoft Windows and the Mac OS X versions, include network operation features. Fundamentally, a network operating system (NOS) is a device-level OS that has been specifically designed to implement and maintain networks. In this installment of IT World, we'll explore the types of operating systems and network operating systems, as well as their basic features.The main features of an OS and NOS include:The basic feature of all network operating systems is user and group management. This is also an area that defines security in that your login is associated with an access rights list of what you are allowed to access on the network—including allowing users to access the Internet. It includes the ability to add, remove, and manage user security features, including authentication, authorization, login restrictions, and access control to data and resources on the network.There are two main NOS types: peer-to-peer and client server. The distinction is pretty obvious. With clientserver, the relationship of networked devices is that the server provides the client with network resources. In a peer-to-peer network, there is no server—just clients that are peers. The peer can share resources with others via the network connection. In a way, you could say that it's a dynamic client-server network in that whenever I provide resources to you, I'm the server and you're the client (and vice-versa).A peer-to-peer NOS provides many of the same resources and services as does a client-server NOS. In a suitable smaller-scaled environment, peer-to-peer networks can deliver acceptable performance levels. They are also easy to install and usually inexpensive. The main cons against peer-to-peer networks are they provide fewer services, and the services they do provide are less robust than those in mature, full-featured client-server networks. The performance of peer-to-peer networks decreases as use increases, especially as the network grows. Maintenance can also be more difficult as there are no methods of centralized management. There can be many separate “servers” to manage rather than one centralized server. In addition, people may have the rights to change their computer's configuration making their use as a “server” unreliable.With a client-server network, the main pro is that clients can be “thin.” Easier to maintain, thin clients do not need all the computing horsepower as the server, especially since the major part of the computing and file storage is at the server. Less computing power, less memory, fewer applications are hallmarks of a thin client. In a typical thin-client environment, only the keystrokes, mouse movements, and screen updates travel across the network connection. However, clients don't have to be thin as long as they can appear to be. It seems to be a step backward when you put your powerful computer into “dumb terminal” mode in order to communicate with the server. In a sense, we've come full circle—back to the mainframe to terminal time-sharing environment!Historically, a true NOS ran on top of the computer's OS, much like the earlier versions of Microsoft Windows that ran on top of the disk operating system (DOS) as seen in Table 1. Remember when the PC booted into DOS, and then DOS called and booted the graphical user interface (GUI or Windows)?The operating system acts as an interface between an application and the hardware, as seen in Figure 1. The user interacts with the hardware via the application. The OS provides a set of services which simplifies the operation and development of applications. Running an application or executing a program involves the creation of a process by the OS's kernel. The kernel is the level of an OS that contains the system commands. The kernel also contains all device drivers, memory management routines, scheduler, and system calls. The kernel creates a process by assigning memory and other resources, establishing a priority for the process, loading program code into memory, and executing the program. The program or application then interacts with the user and/or other devices and performs its intended function.Obviously, an application program provides a particular service beyond what the OS provides. Application programs include word processors such as Microsoft Word, spreadsheets such as Excel, e-mail programs such as Eudora or Microsoft Exchange, presentation programs such as HyperStudio or PowerPoint, Web browsers such as Netscape Navigator, and many others.Overall, operating systems can be grouped according to hardware platform, functionality, and their intended purpose. There are operating systems for all manner of computing platforms and uses.One of the most important aspects of a NOS is its directory, providing information on resources available via the network—including files, users, printers, data sources, and applications. The cool part is that servers communicate with other networked servers to share directory information, making available and keeping current with all the active services on the extended network. The ITU-T standard for directories is called X.500. It allows applications such as e-mail to access information to be centralized on one server or distributed among several servers. The X.500 standard for directory services provides, for example, the means to consolidate e-mail directory information through central servers situated at strategic points on the network. The X.500 servers then share or exchange e-mail directory information to keep all users current. With X.500, any UNIX, MS-Windows, MAC workstation, or client can be listed and make use of the directory. The now-defunct Banyan VINES was the first NOS to have directory services like this, but was pooh-poohed by the industry behemoths at the time. Now, most everyone relies on directory services.Is cloud computing the wave of the future? Windows Azure is one of the cloud computing operating systems—or “cloud services” as Microsoft calls it. Think of it as computing services from a utility. You “log on” to the cloud or network and have access to all the available resources. This includes computation power, data services, printing services, well, just about anything you can think of. You don't need to have the applications or computer power on your machine or terminal; they're in the cloud. Like a utility, you pay for only what you use or subscribe to. As the Internet and your connection to it increases in bandwidth and response time, it matters less where the computing is actually done.This is similar to Google's upcoming Chrome OS, a web operating system. Designed to work exclusively with web applications, it's aimed at users that spend most of their computer time surfing the Internet. Laptops, therefore, become the thinnest clients with no need for hard drives, CD/DVD drives, or other power-consuming hardware. Battery life and connection time is therefore increased, allowing us to spend more time online. Yippee! Well, not really. I spend enough time online as it is, but it will help the virtual society grow and flourish. At least the boundary lines between virtual reality and reality aren't as gray as the lines between an OS and NOS—or are they?