File transfer protocols Essay Example for Free

File transfer protocols Essay File transfer is the process of transmitting complete files from one computer to another. To achieve a successful transfer, both sending and receiving devices must establish the protocol (set of rules) by which they will communicate. Various protocols are commonly recognised but they vary in their performance and suitability for particular tasks. Some examples are described below. FTP File Transfer Protocol (FTP) allows large amounts of information to be transferred between computers very conveniently. FTP originated with the UNIX operating system and was to transfer files of all types between remote computer systems. A popular application of is for the uploading and downloading of Web site pages. ASCII This is only appropriate for text files, which contain no control characters. Thus it cannot be used to transfer files produced with a word processor, spreadsheet or graphics package. Neither can it transfer command (COM) or executable (EXE) files, or files in compressed (ZIP, for ex) form. Apart from this, the protocol is not good at controlling errors. Xmodem A file-transfer protocol used for asynchronous communications. It is commonly used in communications packages. The Xmodem protocol transfers data in blocks of 128 bytes, giving each transmitted frame a sequential block number; this number is used to identify the location of errors. A checksum (see Block Check Characters) is attached to each block to check transmission errors. Its ability to find and correct errors makes it suitable for the transfer of files, which must retain their integrity, such as program files. Zmodem Is one of the most advanced protocols, being much faster than Xmodem. Its error correction controls are absolutely reliable. CCITT V42bis. This protocol includes a data compression (through encoding, data is reduced in volume) technique and error detection and correction. Both the sending and receiving modem must possess the error correction facility. Implementation Top Down implementation is one way to achieve implementation of software. This involves coding, testing and debugging each module in turn starting with the top level and working down. This means that each new module fits naturally into the module hierarchy. Starting with the control module i. e. Songs; we code, test, debug and retest it etc. This involves calling Set_Up, Process_Votes and Menu, even though these have not yet been coded. Therefore, we create test stubs to ensure that we can test that the Songs module achieves its function. A Test Stub can simply consist of a statement which will display a message showing that the module has been called correctly, or it may consist of a set of statements which simulate the operation of the module in order to return results to the calling module which is under test. At this stage, the program would look like this Program Songs(Input, Output); Plan should be constructed which indicates an incremental development where each module is to be tested in turn, with appropriate test stubs. For each test, expected results should be noted. A Test Log indicates the results of each test run, whether the test is successful or not. If a test is successful and an error has been discovered then the code should be redesigned, amending pseudocode, recoded and retested. Total_Votes In 3. Narrative Description No Module Description 0 Songs Calls Set_up to set up the table, Process_Votes to read votes from the keyboard and Menu to offer options. 1 Set_Up Calls, in turn, Initialise and Load. 2 Process_Votes Reads in votes from the keyboard and increment the appropriate total. 3 Menu Displays menu, accept choice, and call appropriate modules Display or Report 1. 1 Initialise Initialises all entries of the table. 1. 2 Load Reads in the song details from the file and store in a table. 3. 1 Display Displays the number of votes for an input song. 3. 2 Report. Displays the percentage votes for each song. Subprogram. Stepwise Refinement is a process which takes an initial solution to a problem and refines each step until the algorithm is detailed enough for implementation in a suitable high level language. Level 0 Level 1 Level 2   Since programs can become very large and consequently can involve a number of programmers in the development process, it is useful to have a way of splitting up a program into components each of which can be handled separately. This uses a facility called a subprogram. The process of developing a solution as a collection of interconnected subprograms is called Modular Design. Each module has a single complete function and is implemented using the particular subprogram facilities of suitable high level languages. Pseudocode Each programmer can be allocated a separate module / group of modules to work on in isolation. Each module may call lower level modules. A module may be called from more than one other nodule. Programmer A Programmer B Programmer C   A subprogram is a way of grouping together statements which havea common purpose i. e.all statements help to produce a singleobjective. They allow a natural way of dividing an algorithm intoparts where each individual part can be handled separately andcarries out a single task for the problem. This is a natural extension of the philosophy of stepwise refinement where subprograms may be executed in sequence, be part of a selection or be iterated. A subprogram can be thought of as a small program which can carry out any of the functions of a program. A subprogram is called when it is needed, executed and then the processor returns control to the place where the subprogram was called from. Although subprograms can be developed in isolation, by separate programmers, they have to communicate with other subprograms in the system, particularly, with regard to the data which is shared between a subprogram and its caller. Subprograms communicate with the calling program through a well-defined interface which specifies the data being shared between the calling program and the subprogram. The interface allows the processor to suspend the calling program, transfer control to the subprogram and pass any data specified down to the subprogram. Once the subprogram finishes executing, then the interface communicates any required data changes back to the calling program and returns control to the main program at the line immediately after the subprogram call. A subprogram may be called any number of times within a program. The subprogram heading specifies the subprogram name, the type of each piece of data which is expected and the nature of each piece of data, i. e. whether it is simply to be passed down in order to aid the function of the subprogram or whether it. is a value which is being changed or calculated as part of the function of the subprogram. A piece of data which is being passed to/from a subprogram is called a parameter. A parameter has an associated data type e. g. integer. A parameter can be specified as either value (reference) or variable depending on whether the subprogram is allowed to change the parameter value permanently. This allows the program to protect certain parameter values against accidental change. e. g. Subprogram Add_Nos(Numl, Num2 : Integer; Var Result : Integer) Here Num1 Num2 are value parameters therefore the subprogram will expect to receive input values for each of these parameters every time the subprogram is called. These values can be supplied either as variable values or constants. e. g. Call Add_Nos Using Numi, Num2, Result Call Add_Nos Using 5, 6, Result Result is a variable parameter and therefore the subprogram will change/calculate a value which is to be returned to the calling program. Thus the subprogram will expect to receive the name of a variable of the required type where it is to store the returning value. The subprogram call must contain a data value (of a suitable type) for each corresponding parameter in the subprogram heading. When a subprogram is called the data values in the subprogram call (actual parameters) are passed down to the corresponding data values in the subprogram heading (formal parameters). Since a subprogram can be used more than once within a system then it is necessary to allow different names to be used for formal and actual parameters. e. g. Call Add Nos Using Nol, No2, Result Exercise Design an algorithm which will input 2 integer values, display a menu, read a choice and call an appropriate subprogram to add, subtract, divide, multiply these numbers returning the result of the calculation. The Systems Development Cycle System Testing and Implementation Implementation is the stage where the system design is converted to a working system. This is arguably the most important stage of systems development, since a well designed system which is poorly implemented may. fail, whereas a badly designed system which is well implemented may survive. The key to success is in planning for the implementation early in the design process, the main issues to be considered being :- ? site preparation ? hardware/software procurement ? system testing ? user education and training ? changeover ? Outcome 6 of the Systems Development 2 Unit includes system testing, user education and training and changeover, and these topics are outlined in the following sections. System Testing System testing ensures that the system works accurately and efficiently before it is put into live operation. All clerical and computer procedures are tested. Programs are tested individually and then in groups to verify program linkages. Hardware and communications links are also tested. The final stages of testing are the system trials involving user staff, this also being part of user training. Test data for program testing should follow the guidelines of the Software Engineering Unit. System trials should be carried out in an operational environment, using test data from earlier time periods, or even parallel runs with the old system using live data. The types of tests which should be considered for a system are as follows Inputs- form design (ease of use) data transmission input validation and correction Files- empty files addition/deletion/update access controls/security Outputs- output documents/screens recipient understanding ComputerProcedures- program linkages operating procedures recovery and security timings Clerical Procedures- effects on other systems user understanding error correction timings User Education and Training The success of a new system is very much dependent on the attitudes of the people who use it. Education aims to convince users of the benefits of the system to themselves, their department and the company. Management policy regarding re-deployment ,redundancy etc should be clearly stated. while casualties may result from the introduction of new systems, there are also many positive aspects such as ? reducing very boring clerical activities? upgrading of skills ? ? improving job satisfaction ? d? Education, therefore, complements and is a pre-requisite to training. e? Training enables users to understand their role in the new system, and prepares them for their own specific tasks. The users and people affected by the new system must be trained, and arrangements for release from normal duties must be made. Many different methods of training are available including f? formal lectures a? ? discussions/seminars ? computer-assisted training (tutorial packages,help facilities etc) ? user manuals/quick reference guides ? Usually a combination of methods will be chosen with an emphasis on hands on experience. Timing and pacing of training is also important, too early and users may forget aspects or become disinterested, too late and users will become flustered. Training does not cease with implementation, but must meet needs of new staff and system changes. changeover Implementation may involve change from a manual system to a computer system or from one computer system to an other. In both cases the system data and procedures require to be changed. A major task prior to changeover is file conversion. The data in existing clerical or computer files must be converted to the new system files, and this can be a massive task involving the preparation of thousands of records using data from a number of different sources. It is also essential that the new system data is accurate or system credibility will suffer. Changeover is the period from the start of live running to the time the DP department withdraws system support, other than for maintenance. ? Methods of changeover depend on the particular circumstances of a system development project, and fall into the following categoriesdirect changeover ? parallel running? pilot running ? Direct changeover is the complete replacement of the old system by the new in one move. This may be appropriate when system testing has been exhaustive and users have complete confidence in the new system. Also if the system is totally new and there is no comparable old system, or if there is no staff available to operate a parallel run. Parallel running is the method of processing current data using the old and new systems at the same time, then cross-checking results. It may be seen as an extension of testing, and generates confidence among users, however it places a strain on resources. Pilot running may be used in two ways, either in retrospective running of data from previous time periods, after results have been obtained from the old system, or as a staged changeover where a complete logical section of records is chosen and committed to the new system without parallel operation in the old system. when this section is operating satisfactorily the remaining records are transferred. The end of the changeover is signified by user acceptance of the new system. System performance must meet acceptance criteria previously established. Acceptance criteria would relate to factors such as response times, degree of reliability and tolerable error rate. Test Narrative An incremental development approach will be taken. This will involve each module being designed, coded and tested in turn before moving onto the next. This Top Down approach will allow newly written modules to fit naturally into the module hierarchy. Test stubs will be used to simulate the calling of modules which have not yet been coded. Menu Tests would be continued until all tests completed amendments logged and retests completed satisfactorily . Interviewing Interviewing is better than questionnaires for gathering certain information because you could see the interviewee facial expressions. Also one other advantage is that in an interview you can hear the response of the interviewee has to say. One disadvantage about the interviewing is that it takes long of time. But couple of things you would gain from the interview is that you would be gaining the trust in the interviewee, you would be making the interviewee feel important and it shows that you are interested. They will be able to give to the interviewer information that will be of value for the new system. If the interviewee says that the current system is unsatisfactory then you will need to look at the weak point of the current system and making sure that you dont make the new system have similar problem as the current system. Interviewing is factual and you would get the answers to know questions right away. Some other problems in a interview is that there isnt enough detail to know whats happening in the job and needs to meet the interviewee to observe the body language. Interview is long but an questionnaire is quicker but it has less information. Questionnaire Questionnaires are useful when you are gathering a small amount of information from a large number of people. But if you are getting a lot of information from people then it will take you a lot of time to process the whole information which will take a lot of time. A questionnaire is quick if it is a small amount of information, it is simple but a questionnaire has a slow response. A questionnaire is quicker than an interview. Questionnaire is one of the best resources to get information from. Like documents they do have information but documents it may not be regularly updated. Examination of documents is factual and has a lot of information but depending on what you are looking for. In a questionnaires you wouldnt get as much information as examination as documents. Some problems in a questionnaire is the person may not be as honest as the person would be in a interview, an questionnaire is too long and boring. Questionnaire is easy to analyze. Study of forms and documents. Procedures manual is an already made source of information so all you got to do is read the information. Examination of documents is the quicker than interviewing and Questionnaires. Records are all factual like all the rest but one of the main things may be bad about is that the information may not be as accurate as a Questionnaire and Interviewing . Input / Formatting / Validation MS Access Input Formatting Symbol. Meaning (space) Display spaces as literal characters.. ABC Display anything inside quotation marks as literal characters. ! Force left alignment instead of right alignment   Fill available space with the next character. Auto Data Types in MS Access Data Type Explanation of Data Type Text Alphanumeric Max 255 alpha-numeric, characters, 50 is the default Memo Alphanumeric Max 64000 alphanumeric characters Number Numeric Numeric values for calculation Five formats Date/Time Dates/Tinies Five date and three time formats Currency Monetary Monetary values up to 15 digits, plus decimal places AutoNumber Unique Value for each record Sequential numbers automatically inserted by access Yes/No Boolean Logical values in Yes/No, on/of for True/False formats OLE Pictures/Graphs or other OLE Object Linking and Embedding items, graphics or binary Hyperlink. Decimal placeholder and thousand date and time separators. (The actual character used depends on the settings in the Regional Settings section of the Windows Control Panel). Causes all characters to be converted to lowercase. Causes all characters to be converted to uppercase. ! Causes the input mask to display from right to left, rather than from left to right, when characters on the left side of the input mask are optional. Characters typed into the mask always fill it from left to right You can include the exclamation point anywhere in the input mask \ Causes the character that follows to be displayed as the literal character. If you create a validation rule for a field Microsoft Access doesnt normally allow a Null value to be stored in the field. If you want to allow a Null value add Is Null to the validation rule as in 8 Or Is Null. Query criteria in MS Access Field Data Field data finds an exact match for a field Smith will find smith Access is not case sensitive wildcards Two wildcards can be used to replace field characters. any single character 0 any characterds values greater than or equal to 01/12/92 Compound Criteria The user can specify more than one example and criteria at any one time.