Characteristics of the Term Multitasking and Its Types - Essay Example

MULTITASKING Introduction Prior to the introduction of multitasking, the early programs used to run in batch mode on computers, i.e. once a particular program was initiated it continued execution till the batch of instructions was completed. User interference was not allowed or possible while the batch was running, since two programs could not run simultaneously. However, after continuous research a new function, called multitasking was developed whereby multiple programming could be facilitated on the same machine. It enabled the programmers / users to run multiple tasks at the same time, by allowing two or more programs complete access to the system, in the batch mode. This facility helped in improving efficiency1. Meaning With respect to computing, the term Multitasking refers to a process which facilitates undertaking of several tasks or processes at the same time, by enabling sharing of certain common processing resources such as the CPU for instance. Multitasking refers to the running of multiple independent computer programs on the same computer; giving the appearance that it is performing the tasks at the same time. The kernel of the operating system is responsible for that activity. Kernel In computing terms, kernel acts as a vital constituent of a majority of computer operating systems. Its key responsibilities include administering the operating systems resources, for instance, setting up and ensuring adequate interaction between the hardware as well as software components. One of the fundamental functions of the kernel includes administering the computer’s resources and permitting other programs to run and use these resources. Characteristically, the resources comprises of: the CPU (central processing unit), the memory of the system, as well as the I/O (Input / Output) devices. The kernel of a multitasking operating system alternatively referred to as KMOS, facilitates four basic activities2: The inter process communication and synchronization Interrupt management Process management and Booting and startup of the initial system configuration The multitasking characteristic entails the necessity to sustain simultaneous implementation of processes and to offer services for inter process communication and synchronization. Other less noticeable yet equally significant functions of the kernel comprises of management of interrupts and basic process scheduling. Furthermore, apart from the above listed functions, the kernel typically manages some ancillary actions essential for offering the services which are observable by the other Operating System layers. Examples of such internal kernel operations include system time-base management and processor allocation by means of basic process scheduling. Multitasking can be classified into two major forms: Cooperative multitasking and Preemptive Multitasking. In Cooperative multitasking, the tasks must from time to time and momentarily relinquish its authority willingly while in preemptive multitasking the tasks need not be concerned about such an aspect, since context switching is indiscernible to tasks. The operating system intangibly anticipates each and every task whenever it desires to transfer power to another, and afterward reinstates the first task without even acknowledging the fact that it was suspended momentarily. Figure: Time-slicing allows concurrent tasks in the same host3 Executing simultaneous tasks in a single host – a process popularly referred to as multitasking; is a significant characteristic of contemporary operating systems. Since computers in present times, are relatively swift, multitasking is not apparent to the user, who is under the belief, that a host can run diverse applications concomitantly with little or no noticeable impact on one or the other4. One significant incentive for multitasking, apart from the inherent function requirement, is equality. On a single host, the substitute for multitasking would be to carry out tasks successively one after the other. This may be regarded as unjust if one task has a very extensive finishing time, shutting out the other tasks. The influence of simultaneous performance / execution of each task is, hence, without a doubt larger (given that the computer is exchanging among them). However, a brief task may come to an end more rapidly - since it is permitted to commence earlier - if the tasks are synchronized. For instance: In the absence of multitasking, a user would have to wait for long hours for a particular task to be duly executed and completed, prior to proceed towards achievement of other tasks on hand, which may, on the other hand, take up much of one’s time. Figure: Multitasking in the same host contributes to fairness Multitasking consists of a specific similarity to the communication of messages. Similar to a task, a message cannot be regarded as constructive if it is not obtained in its complete form, or as a whole. In several applications, the capability differs erratically or arbitrarily with time. An untoward outcome of such variations is overcrowding or jamming, which is a temporary scarcity of resources on account of momentarily high influx of workload. Overcrowding or congestion or jamming occurs on account of several tasks competing for a hosts computational resources, or from larger than usual messages competing for a networks communication resources. It significantly affects the performance unswervingly by impeding task commencement, or increasing task achievement time, or with respect to communications, postponing the moment of a message’s appearance at its final goal or destination. The most significant factor regarding overcrowding is that processing and communication resources have to be adjusted in accordance with the climax load, meaning thereby that more often than not, they are underutilized5. Micro-Kernels v/s Monolithic Kernels Micro Kernels Kernels can be generally categorized into two main types: monolithic kernels and micro kernels. Monolithic kernels comprise of all kernel functionality in one single /large package while Micro kernels comprise of only those functionalities which have to unconditionally be a part of the kernel leaving subsidiary utility to be laden as modules whenever need be. Figure: Micro kernel Architecture The key benefit of the microkernel is that the slighter set of software alters less frequently, and therefore is relatively steadier. The majority of amendments to kernel code are auxiliary roles, such as device drives which must transform as new hardware is released. The microkernel facilitates such a transformation by allowing it to modernize a driver without having to restore the microkernel itself. The drawback of the microkernel system, however, include the performance appraisal taken when module loads take place, as well as the safety hazards of interpretation of specific codes from a file and loading it into the kernel. Micro kernels also endure the overhead of replicating messages, which is costly in contrast to passing pointers to static buffers in a common kernel-wide address space. As a result the primary benefit of a monolithic kernel is its remarkable performance6. Monolithic Kernels Monolithic kernels offer all of their critical functionality through a distinct procedure. As a result, they are considerably bigger as compared to the micro kernels. Characteristically targeted for explicit hardware, monolithic kernels cooperate openly with the hardware, so they can be optimized without difficulty, as opposed to the microkernel operating systems. It is on account of this factor, that monolithic kernels are not effortlessly transportable. Examples of monolithic kernel operating systems include Linux, MacOS and DOS7. Figure: Monolithic Kernel Architecture Several major computer systems including Linux and Microsoft, which have used both the types of kernels, prove that speed is not one of the major determinants of the type of kernel to be used. The supplementary all-embracing compatibility and transportability preordains them for different applications reaching from embedded to desktop systems. They are relatively simpler to preserve as compared to their monolithic counterparts. Micro kernels offer undeveloped operating system functionality, depending on other modules to carry out definite tasks, consequently shifting several archetypal operating system services into user space. This authorizes numerous services to be resumed or reconfigured without restarting the whole operating system. Micro kernels offer protection, since services functioning at the user level have limited admittance to system resources. Micro kernels can be modified and ported to other hardware more effortlessly than monolithic kernels. Hence, micro kernels are preferable as compared to monolithic kernels. References: Walters, E. G., (2001). The Essential Guide to Computing, Pp. 143 - 144 Kumar, S., (2004).Encyclopedia of Operating System, Anmol Publications, Pp. 666 Messerschmitt, D. G. (1999). Understanding Networked Applications, Morgan Kaufmann, Pp. 471 - 473 Carling, M., Degler, S., Dennis, J., (1999). Linux System Administration, Sams Publishing, Pp. 101 - 103 Null, L., Lobur, J., (2006). The Essentials of Computer Organization and Structure, Jones & Bartlett Publishers, Pp. 414 - 416 Read More