Client/Server Architecture
1. Do you think the problems faced by Hures is unique to them? Why or why not?
The Client-server problem that Hures has encountered is not unique to them. Yes, Client-server computing had a positive impact on organizations, but, as with all new technologies, there are many challenges and questions with hard-to-find answers.
Desktop Management Task Force (DMTF), a standards organization founded in 1992, is creating the most important set of standards for network and systems management since the initial version of simple network management protocol (SNMP) was released in 1988. DMTF is tackling the significant problems involved with managing hardware and software of PC desktops and servers. Its specs are complete enough that users can now require conformance from vendors and look for management applications that incorporate the specifications in their architecture.
Among the vendors that have supported DMTF since it was launched in 1992 are Intel, Microsoft, Novell, Digital, Hewlett Packard, IBM, SunSoft and SynOptics. Since then many others have joined, including AST Research, Compaq, Dell, Symantec and Apple, and hundreds of hardware and software vendors have pledged to implement the DMTF specifications.
It may seem strange that this large group of vendors has decided to work together to make your life easier; historically, the opposite has been true. But the DMTF members aren’t altruistic–they’re cooperating because they believe that by easing the management burden, customers will move toward client-server computing faster.
To make client-sever networks run smoothly, we will have to be able to manage the desktop. The DMTF members are developing MIFs and helping developers incorporate DMI into products through workshops and publications. The work of the DMTF should make 1995 the year it became possible to manage desktops in a cohesive and simplified manner. Will the DMTF be a success? Absolutely.
2. Suggest an alternative technology that could be used to overcome the problems faced by Hures with their Client/Server technology.
The lack of effective remote management for desktops is a major barrier to widespread adoption of client-server computing. The PC has become the ultimate open system. The complexity problem is even worse at the workgroup and enterprise levels. Workgroups often contain workstations from multiple vendors, some of which have attached printers, others have CDROM drives and each has a different hard-drive capacity and runs different software.
Today, managing these environments requires an arsenal of separate vendor-developed utilities that for the most part cannot be invoked remotely. What is needed is a standardized management agent that the various components can share and a uniform and easily extended way to format management information and commands.
DMTF specs will help save time and money. They will maintain consistency throughout a network and will free managers to work on the kinds of activities that push users forward rather than on the tedious housekeeping tasks that hold them back.
The DMTF has addressed a problem that SNMP never defined: a set of open APIs for accessing a common agent. DMTF’s agent is called the Desktop Management Interface and it defines how management applications request services from an agent through a Management Interface (MI). It also defines how component providers can provide routines that integrate into the DMI agent structure through the Component Interface (CI). These definitions result in an open agent platform that allows vendors to share a common management interface for PCs or workstations.
The DMI is independent of operating systems and protocols. Think of it as an interpreter: It takes requests from management applications and translates them through the DMVs Service Layer, so the component can give the application the information it needs. The DMI also allows components to send notifications through the Service Layer to any application that is registered to receive this type of information.
The DMI enables component-specific agent modules to be executed in response to management commands that come through the DMI interface. It will be possible to remotely invoke disk diagnostics or test routines on specific boards in a machine in a standardized way.
In the DMTF specification, the definitions of management data are called Management Information Files (MIFs), and they resemble SNMP’s Management Information Bases (MIBs). A MIF is a text file that contains information about the component, its manufacturer, its configuration and any other information that will help manage the device.
To assure that the information stored in a MIF remains the same, regardless of which vendor creates the product, the DMTF is publishing standard MIF structures for different components. Having a standard way to specify desktop and server configuration information is a key breakthrough, and it will enable us to automatically collect a full inventory of all hardware and software. This will have the same impact on systems management that SNMP’s auto-topology and discovery capabilities have had on network management.
DMTF’s work is carried out in working groups, which are organized around specific types of devices. Each group is establishing a definition and structure for the MIF information to be stored on its components.
The PC MIF contains information on the CPU, operating system, motherboard, audio, video, BIOS, I/Os, co-processor, power source and expansion slots.
3. One of the suggestions proposed by Hures IS department is the use of Intranet Webs. Examine the pros and cons of such an idea. An intranet is a framework for creating and sharing corporate knowledge, the emphasis being on the word framework. Based on Internet standards and tools, an intranet focuses on content sharing within a limited and well-defined group.
Benefits of intranets. There are a number of major advantages in having an intranet, such as: ease of use; single intuitive user interface (namely the browser); widespread industry support; and ability to link up legacy systems (both applications and databases).
Commentators acknowledge that intranet equipment will outsell that used for the Internet by a considerable margin. This begs the question of why there has been such a huge take-up of intranets. The reasons why intranets are so successful are: (1) Platform independence—web technologies allow a whole range of different platforms to be linked by a common interface. (2) Quick wins—an intranet can be set up very cheaply, but if it is well designed it can make a big impact upon internal communications. Something as simple as putting the phone book onto the intranet can save a large organisation thousands of pounds on reprographics, and the end result is a list that is always up to date. There is no longer the danger of some copies being up to date, and others being six months out of date. And (3) Cost savings—there are a number of potential cost savings. These include user support. On an intranet, all the processing and applications sit on the server. It is only necessary to update the software on the server, instead of on each individual user’s pc. Another area of cost savings would be in network management, since there would be a move from a multi-protocol network, to a single protocol network, namely TCP/IP.
Cons of Intranets. There are also disadvantages: (1) Management could lose control of the material provided in the Intranet. (2) There could be easy security concerns with who accessess the intranet, plus abuse of the intranet by users. And (3) Intranets may cause “information overload”, delivering too much to handle.
According to an article in in May 1997, ‘Web-based technologies provide a common user interface across disparate platforms, enabling the development of cross-platform applications, the integration of existing legacy applications and databases, and enhanced communications and information sharing among the intranet users’ (1997).
In order to achieve the business benefits which an intranet can undoubtedly bring, you have got to address the cultural issues. You have to face headon the question of how you can create a culture within your organisation which will encourage all your staff to use the new technology and to share information. What you should be aiming for is a learning organisation, one that is able to learn from experience, and one which doesn’t continually reinvent the wheel. That means having the confidence to try things out, being willing to fail in order to learn how to succeed, finding out what works and what doesn’t. This isn’t going to be easy—who wants to share their failures with work colleagues? But, if they do so, then hopefully together employees can help each other along the road to success. The collective brainpower of an organisation is clearly far more powerful than that of any individual. If everyone is working together, then they will bring together collectively a number of differ ent thoughts, ideas and perspectives. To achieve this, there must be an information sharing culture.
I think that one important factor is to ensure that employees feel that it is worthwhile sharing information, or providing feedback and suggestions about the intranet. This doesn’t have to be about monetary rewards. Rather, it is about demonstrating that you value their efforts. Show them that you have taken on board their comments and suggestions, and that you have acted upon them. Or, if for some reason the comments are impractical to implement or don’t fit in with the overall strategy; take the time and effort to explain why.
4. Do you think the popularity of Intranets and the Internet pose threats to traditional Client/server systems?
Just because people have discovered intranets doesn’t mean all companies should get one simply because their competitors have got one. Information Technology departments must enter into a dialogue with their firms′ senior management in order to try and understand what the key issues affecting the business are, because these should be what drives development, not IT issues. Unless this point is fully understood, businesses run the risk of failing to exploit the strategic and competitive advantages which are potentially available to them through investing in intranet technologies.
It is easy to envisage the chief executive of a company going along to lunch with his counterpart in a similar organisation. When he has lunch, his counterpart sings the praises of the wonderful new intranet that they have just installed in their organisation. On his return from lunch, the chief executive calls in the IT director and says, ‘We must get one of those’. And so, an intranet is duly installed, and the chief executive can now look his peers in the eye and say that his company also has an intranet. How many times do people find out that their competitors have a particular piece of software, so they must get it too? What is important, of course, is not whether or not an organisation has an intranet, but rather to what uses it is being put. Over and over again, when you ask people what their intranet contains, you find a remarkable unanimity amongst organisations, where the intranet basically contains the telephone directory and the menu for the staff restaurant, but nothing of any substance. I was amused to learn that this type of intranet has a name— they have been dubbed ‘emptynets’ (1999).
It is easy to mock such things, but it very much depends upon the future development plans that an organisation has for its intranet. If a company just puts up the phone directory and the menu, and thinks that it now has a fully-fledged intranet, then something is wrong. But if, however, that organisation has put up those applications deliberately as a ‘taster’ of what an intranet looks like, and as a means of getting people interested in the idea of an intranet then clearly there is the potential to ask the users what they would like to see on the intranet, and how they would like it to develop. It is vitally important for companies to realize that an intranet is something dynamic. It has the potential to grow into something that becomes a natural part of office life, just like using the photocopier or the fax machine. Once people grasp that there is a developmental process involved, then the intranet could lead on to an extranet, and indeed to electronic commerce (also known as e-commerce).
Credit:ivythesis.typepad.com
0 comments:
Post a Comment