Harvey M. Deitel. An Introduction to Operating Systems. Addison-Wesley Publishing Company, Reading, Massachusetts, 1990.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....field of device driver design. The next chapter starts with an overview on the internals, the interactions, and the objectives of a device driver within an computer system. Anatomy of a Device Driver Device drivers play only a secondary role in operating system course literature (for example [19, 106, 84, 64, 102, 70]) The literature mostly focuses on processes, scheduling, and memory management. An upcoming topic is security and distributed computing. The access to external devices is described in literature only from a very high level. The device drivers are just seen as modules which provide access to ....

....combination with the directly assigned work, this chapter provides an overview on the different aspects and view points of driver design. 3.1 Books on Device Driver Design Device driver architectures play only a secondary role in the literature on foundations in operating systems. Examples are [19, 106, 84, 64, 102, 70] which discuss more the topics on process and memory management, networks and security. Device access is only a brief section. The hardware is mostly described as a device controller which is accessible by the system bus. Books on programming in the different operating system environments like ....

[Article contains additional citation context not shown here]

Harvey M. Deitel. An Introduction to Operating Systems. Addison-Wesley Publishing Company, Reading, Massachusetts, 1990.

.... Therefore, many scheduling policies concentrated on minimizing seek time such as Shortest Seek Time first (SSTF) SCAN, LOOK, and V(R) and those concentrated on minimizing rotational latency such as Shortest Latency Time First (SLTF) have been proposed in order to achieve higher performance [16][17] 18] 19] In other words, these policies attempt to service I O requests with the minimum mechanical motion. However, they are less concerned about each request individually, which is what our policy does. As a consequence, the problem similar to that when using traditional CPU scheduling ....

H. Deitel, An introduction to operating systems (Addison-Wesley, 1990).

....the makespan) Both metrics have the disadvantage that they are hijacked by long running jobs. In the extreme case when one of the jobs has an infinite loop, the metric is completely blind to shabby treatment of the other jobs. Perhaps this 46 is why standard operating system texts such as [PS85, Dei90, Tan92] completely ignore traditional scheduling theory. We now introduce a different metric that explicitly recognizes the requirements of interactive jobs. The motivating principle behind the metric is: No job should be ignored for a long time. Here, long time is defined as anything greater ....

H.M. Deitel. An Introduction to Operating Systems. Addison-Wesley, Reading, MA, 1990.

....throughput provides the most overall benefit. Chapter 2. Background 15 2.1.2 Uniprocessor Results To understand the development of multiprocessor scheduling algorithms, we will first consider some well known results concerning single processor scheduling. A further discussion can be found in [Dei90] Scheduling policies for a single processor can be classified by how much information is available to the scheduler. If the service demands of all jobs are known explicitly, then Shortest Job First (SJF) running the jobs in increasing order of service demand, is the best policy for minimizing ....

Harvey M. Deitel. An Introduction to Operating Systems. Addison-Wesley, Boston, 1990. Bibliography 85

....The task view of a program P = T; S consists of ffl a set of T components of type task; ffl a set S of connectors of type spawns. A task or thread is a subprogram or a piece of code that can execute concurrently sharing the address space with other tasks within a program (see for example [15]) A spawns connector connects two tasks where the first is the creator (father) of the second task (child) The task view represents a program s possible multiple threads of control and their relations. This is a slightly different view with respect to that of the invocation connector in the ....

H. M. Deitel, "An Introduction to Operating Systems ", Addison-Wesley, 1990.

....The task view of a program P = T; S consists of a set T of components of type task and a set S of connectors of type spawns. A task or thread is a subprogram or a piece of code that can be executed concurrently sharing the address space with other tasks within a program (see for example [14]) A spawns connector links two tasks, in which the first is the creator (father) of the second task (child) The task view of a program represents possible multiple threads of control and their relations. This is a slightly different view with respect to that of the invocation connector in the ....

H. M. Deitel. An Introduction to Operating Systems. Addison-Wesley Publishing Company, Reading, MA, 1990.

....deem you unprepared and remove you. If you cannot allocate about 20 hours a week to work on this class, you will have difficulty. 1 3 Some Additional Resources Some good books for alternative information include: ffl Overviews: Stallings [6] for a qualitative operating systems approach. Deitel [3] provides a good introductory mix of qualitative and quantitative approaches. Maekawa, Oldeheft and Oldeheft [5] provides a more advanced coverage of topics discussed in class. Tanenbaum [9, 8, 7] has written a series of books about operating systems design and implementation. ffl Operating ....

H. Deitel. An Introduction to Operating Systems. Addison Wesley, Reading, MA, 1989.

....may also be acceptable) and understand data structures and assembler language programming. Knowledge of basic probability and calculus is assumed. 2 Some Additional Resources Some good books for alternative information include Stallings [3] for a qualitative operating systems approach. Deitel [1] provides a good introductory mix of qualitative and quantitative approaches. Jain [2] provides a good overview of systems performance analysis techniques (including discrete event simulation) 3 Course Overview This course introduces students to the basic concepts of Operating systems design ....

H. Deitel. An Introduction to Operating Systems. Addison Wesley, Reading, MA, 1989.

....machine dependent if it makes important assumptions about the characteristics of the target machine. One such optimization is global register allocation, and I hope to give an algorithm for this in the thesis. There is a close analogy between register allocation and virtual storage management [Dei84]: in optimal page replacement (register allocation) Bel66] if there are no free pages (registers) then replace (spill) the page (register) that will not be needed again for the longest time. While this analogy works well for local register allocation, using it globally is problematic in a CFG ....

Harvey M. Deitel. An Introduction to Operating Systems. Addison-Wesley, 1984.

....step) ALF [4] Statemate [7] and ProcessWeaver [2] also support a rather limited amount of resource modeling. Related work in software systems Modeling resources has been an important activity in operating systems, where resource models are used as the basis for scheduling (for example, [3], Chapter 6.4) Peripheral devices and parts of code or data that require exclusive access are generally managed quite precisely, but the resource management needed by operating systems is quite different from ours. In particular, operating system resources are far less varied than the resources ....

H. M. Deitel. An Introduction to Operating Systems. Addison-Wesley, Reading, Massachusetts, 1984.

....is done during the virtual address translation. Now, we start with some quick definitions and explanations of virtual memory systems which are the base for our main topic protection . For the complete subject of virtual memory systems , the interested readers are referred to references: 3] and [4]. 2 Virtual Memory Systems Virtual memory concept is based on the ability of each process to have an address space much larger than the address available in physical memory. It provides many facilities such as: ffl The property of placing Many programs in physical memory at the same time ....

Harvey M. Deitel, "An Introduction to Operating Systems", Addison Wesley, 1984.

....made its commercial debut in 1972. The IBM VM system enabled organizations to run both the MVS and CMS operating systems on the same physical machine by virtualizing machine resources. The monolithic service architecture of this system meant that all VM services were executed on the same host [Deitel 89] This structure has influenced many of the virtual machine implementations that followed. In the early 1970 s, virtual machines were adopted by the language community as a substrate for code distribution. These systems used virtual machines both to retain the portability of applications in an ....

H. M. Deitel. An Introduction to Operating Systems. Addison-Wesley, August 1989.

.... throughput and processor utilization, minimizing the total execution time required to execute the tasks comprising a single job, minimizing the mean job response time, minimizing processor idle time, reducing the communication and scheduling overheads, and avoidance of indefinite postponement [18] [48] Previous research efforts in scheduling have emphasized varying objectives. Some 17 of the scheduling policy issues that have been actively researched in the literature include the order in which tasks are assigned to processors. This issue of task ordering can result in two types of ....

H. M. Deitel, An Introduction to Operating Systems, Addison-Wesley, 1984.

....blocks are read and buffered ahead of the time they are needed, has been introduced. Although early work on continuous media storage systems assumed separate buffers for each client, recent work [9, 12] has assumed a global buffer cache, similar to the buffer cache in traditional storage systems [8, 15] where the cached data can be shared among all the clients. The rationale for using a buffer cache for multimedia systems is the same as that for traditional storage systems: namely, the reduction of disk I O. As opposed to the case of traditional systems, however, LRU does not yield a high cache ....

Harvey M. Deitel. An Introduction to Operating Systems. Addison Wesley, 1983.

....programming in order to appreciate fully the above concepts. This paper describes the SR language and discusses its successful use as an environment for concurrent programming in an undergraduate operating systems class. 1 Introduction Standard undergraduate operating systems textbooks, such as [1, 2, 3] cover the important concepts involved in concurrent process synchronization and interprocess communication. These textbooks describe how critical sections must be identified in the code of processes that execute concurrently and share data so that race conditions can be avoided. Mutual exclusion ....

Harvey M. Deitel, An Introduction to Operating Systems, second edition, Addison-Wesley, 1990.

....consider four important aspects of the Synthesis design that depart from common precedents and trends. 17 2.3 Kernel Structure 2.3.1 The Trend from Monolithic to Diffuse Early kernels tended to be large, isolated, monolithic structures that were hard to maintain. IBM s MVS is a classic example [11]. Unix initially embodied the small is beautiful ideal [28] It captured some of the most elegant ideas of its day in a kernel design that, while still monolithic, was small, easy to understand and maintain, and provided a synergistic, productive, and highly portable set of system tools. ....

H.M. Deitel. An Introduction to Operating Systems. Addison-Wesley Publishing Company, second edition, 1989.

....concepts and the difficulties relating to them will be discussed briefly in this section. Synchronization is needed when one process wishes to perform an operation that can only be performed safely if some other process has already taken some action. For example, in the producer consumer problem [Dei90, FM91], a buffer is shared between one or more producer processes and one or more consumer processes. Producer(s) must wait if the buffer is full, and consumer(s) must wait if the buffer is empty. Since the progress of the processes involved depends upon the existence of certain conditions, this ....

....the basic form shown in Figure 1. 1 It should be noted that it is possible to have several process reading the shared data at the same time, or there may be a single process writing the shared data (with no active reading processes) this is the basic premise of the Readers and Writers problem [Dei90, FM91]. Figure 1 Using ME Protocols process X; loop non CS; entry protocol; CS; exit protocol; non CS; end loop; end X; Problems arising from various inadequate solutions to these algorithms (i.e. the implementation of the entry protocol and the exit protocol) include [Nel91] 1) ME violation, ....

[Article contains additional citation context not shown here]

H.M. Deitel. An Introduction to Operating Systems (2nd Edition). Addison-Wesley Publishing Co, Reading, Mass, 1990.

....number may greatly degrade the performance. Therefore, it is desirable to give long jobs longer scheduling slots to reduce the overhead of context switch, but less number of times to access the service so that short jobs will not frequently be blocked by longer ones and may be completed quickly [3]. The lack of flexibility in varying the scheduling slot and in determining the nature of processes may make the conventional coscheduling algorithm not very efficient. The next problem associated with conventional coscheduling is the cache reloading effect. When the processes in the system are ....

H. M. Deitel, An Introduction to Operating Systems, 2nd ed., Addison-Wesley, Massachusetts, 1990.

....the effects of potential optimizations, but also how and why they affect the system. Historically, computer systems have been studied experimentally by subjecting the test system to a known load and capturing static, aggregate performance measures such as throughput, response time, and utilization [7]. Similarly, the performance of application programs often has been assessed via code profiles that reveal the distribution of execution time across the code. Although these application and system performance measures have been effective for systems with only a small number of processors, they do ....

Deitel, H. M. An Introduction to Operating Systems. Addison-Wesley, 1984.

....is not in service its priority increases and at a certain point its priority is so high that it gets the CPU ( 2, 11, 12, 16] static priorities : In this mechanism priorities always remain the same. Very often a mixture between preemptive and nonpreemptive static service strategies can be found [7, 8, 10, 11, 12, 13, 16]. Several authors have developed analytic approximations for priority strategies. In [1] a method is introduced to analyze systems with a pure preemptive or nonpreemptive service strategy. In this case it is assumed that priorities are constant. In [2] analytical solution techniques for different ....

....basic work in the kernel context and all other work be performed by lower level hardware. Then the system kernel is blocked much less time and the system can do more useful work. This approach, where the kernel context performs only the basic work of a system call, is called micro kernel approach [7]. In this approach, the kernel determines only the interrupt number and initiates the corresponding routine in the system context. After that the kernel context is left. Consider, for example, the case of a page fault in the system. In the case of a micro kernel system, the system traps to the ....

[Article contains additional citation context not shown here]

Deitel H.M.: An Introduction to Operating Systems, Addison Wesley, Reading, 1990.

....Table 1) and the data rate of the new stream 5 . The data retrieval time is calculated as sum of two parts, data transfer time and overhead time. Overhead is sum of the time used by service ahead of the arm it jumps back to service the request nearest the outer track and proceeds inward again[2]. 5 Appendix B describes the derivation of these formulas. other disk access activities, head seek time, rotational delay and command overhead. Appendix C describes total overhead time, O total , in more detail. T Interval time of CRAS B total Total buffer memory size D Disk data transfer ....

H.M.Deitel, "An Introduction to Operating Systems ", Addison-Wesley, 1984.

....spaces that only the operating system can change between. Each application is assigned its own address space and cannot reference any data or functions in other address spaces. Multiple address spaces can be provided in many ways, including: base bounds registers, segmentation, and virtual memory[Dei84a, PS85] As an alternative to building costly high performance processors, the proliferation of low cost microprocessors has allowed system designers to build high performance computing systems out of a large number of small, inexpensive systems. Distributed operating systems support such a ....

....concrete classes implementing the signature can provide different strategies while leaving code using the interface unaltered. For example, a FIFO Manager class might define the replacement algorithm to be first in first out, while a NUR Manager class might define it to be a not used recently[Dei84a] policy. Other common operating system policy decisions that benefit from this approach include disk head scheduling, serial line character processing and file system access methods. 4.2.4 Optimization by Subclassing The examples in Section 4.2.2 show how inheritance allows specialization ....

[Article contains additional citation context not shown here]

Harvey M. Deitel. An Introduction to Operating Systems. Addison-Wesley Publishing Company, Reading, Massachusetts, Reading, Massachusetts, 1984.

....machine was the IBM VM system [IBMVM86] which made its commercial debut in 1972. The IBM VM system enabled organizations to run both MVS and CMS on the same physical machine by virtualizing machine resources. As a result of the monolithic structure, all VM services were executed on the same host [Deitel 90] This structure has influenced many of the virtual machine implementations that followed. In the early 1970 s, virtual machines were also adopted by the language community as a substrate for distributed code. These systems used virtual machines both to retain the portability of applications in ....

Deitel, H. M. An introduction to Operating Systems. Addison-Wesley, Reading, MA 1990.

No context found.

Deitel, H.M. An Introduction to Operating Systems. Reading, MA: Addison-Wesley, 1984.

No context found.

Harvey M. Deitel. An Introduction to Operating System. AddisonWesley, 1983.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC