Ibm Scheduling&process Management

  • Uploaded by: api-3700456
  • 0
  • 0
  • June 2020
  • PDF

This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA


Overview

Download & View Ibm Scheduling&process Management as PDF for free.

More details

  • Words: 1,307
  • Pages: 10
SCHEDULING

control program – supervisor, master & job scheduler - efficiently schedule, initiate & supervise the work by the computing system master scheduler – controls the overall operation of the computing systemoperating system combination job scheduler – enters job definitions into the computing system, schedules & then initiates the performance of work under control of the supervisor * The job scheduler is responsible for job management functions. It permits either sequential FIFO scheduling or nonpreemptive priority scheduling. * The control & the direction of processing data is in the supervisor. * The job & master schedulers perform a vital role in scheduling & supervising the performance of work by the computing system.

SCHEDULING * Before any work can be performed by the system, it must be scheduled & initiated by either the master scheduler or job scheduler. * In the MFT & MVT configurations of the OS, the supervisor is capable of directing & controlling the performance of more than one data processing task at a time. 2 configurations of the OS •multiprogramming w/ fixed number of tasks (MFT) •multiprogramming w/ variable number of tasks (MVT)

SCHEDULING Non-Stop Job Processing job – major unit of work done performed by the OS - each job is defined by a series of job control language statements Job Characteristics •defined by series of job statements •consists of one or more steps •Individual job definitions can be placed one behind another

*Any set of data that is processed during a job step must be identified & defined within the definition of the job step using a DD (data definition) statement. * If a series of job step definitions are to be used repeatedly w/ little or no change, a programmer can store & catalog them in a procedure library maintained in direct access storage by the control program. Multiple Job Processing * The steps of a data processing job are logically related to one another to produce a specific end result. Example of a Job consisting of three steps •Translating a source program into an object program •Linkage editing the object program to produce a program suitable for loading into main storage. •Loading & executing the program.

*Multiple-job processing is particularly suited to data processing installations with a high volume of work & a large number of resources. *With an MFT or MVT control program, these same jobs can be performed either sequentially (part A of the figure) or, if enough resources are available, concurrently as shown in part B of the figure. *A set of data in direct access storage can be shared concurrently among several jobs provided it is not changed in any way the jobs that are sharing it.

Concurrent Job Support • Tasks Job definitions & any input data that accompanies them in an input stream, are usually submitted for processing in the form of punched cards.

*Steps Output data ends up inProcessing printed orJobs punched Performed When card form. •The jobs, in punched card form are normally arranged in priority order. •After enough jobs have been accumulated to form a batch, they are transcribed to magnetic tape. •The batch of jobs on the tape is manually scheduled and then processed on the central computing system. •After a batch of output data has been recorded on a tape by the central computing system, it is manually scheduled & converted to printed or punched card form or a combination of the two. •The printed and punched card output is manually sorted into various classes and distributed to the individuals that submitted the jobs.

*To avoid such problems at MFT or MVT installations, operations such as reading job & data cards & printing job output data, are performed by the control program as separate tasks, concurrently with work. * At other MFT & MVT installations the control program can read job definitions & data from one or more job input streams and record job output data on one or more output devices, while initiating & controlling the performance of one or more jobs.

The MFT & MVT Job & Master Schedulers * Job & Master Schedulers control the concurrent processing of job & job support tasks. MFT & MVT job schedulers are divided into: •reader/interpreter – reads jobs & job step definitions from an input stream & places them in the input work queue •initiator/terminator - selects and initiates the job •output writer – reads data from the output work queue & records it on an output device

MFT & MVT master scheduler – serves as a two-way communications link b/w operator & system; used to relay messages from system to operator, to execute commands like starting & stopping job scheduling tasks, log operational info, monitor & control the progress of work, etc. *Any reader/interpreter & output writer tasks that are to be performed at an installation are defined in much the same way as a single step job is defined. Specifications can be respecified when the operator starts the task. MFT & MVT – 15 initiator/terminator tasks to control 15 concurrent jobs MVT – any number of reader/interpreter & output writers MFT – 3 reader/interpreter & 36 output writer tasks

Job Priorities So far as job management is concerned, the main difference between MFT and MVT control program has to do with the way in which priority is assigned and main storage space is allocated. At an MFT or MVT installation, each job that is submitted for processing can be assigned a specific priority relative to other jobs. It can also be assigned to any one of several classes of jobs. When the job definitions are read by the reader/interpreter they are placed in the input work queue in accordance with their assigned class and priority. A separate input queue is maintained for each class assigned to the jobs. Within each input queue, the job definitions are arranged in the order of their priority. Output data produced during a job step can be assigned by the programmer to any one of up to 36 different data output classes defined at the installation. When an output writer task is started it can be assigned to process from one to eight different classes of output. A particular output class may represent such things as the priority of the data, the type of device that may be used to record it, or the location or department to which it is to be sent.

Job Priorities

In an MVT installation, any main storage space not reserved for the control program serves as a pool of storage from which a region is dynamically allocated by the control program to each job step or job support task as it is initiated. The size of the region to be allocated to each job step is specified by the programmer in the job or job step definition. The priority of a job is also specified by the programmer. When an initiator/terminator task is started by the operator, it can be assigned to initiate jobs from one through eight input queues. By classifying jobs and assigning initiator/terminators to initiate specific classes of jobs, it is possible to control the mixture of concurrent jobs; thus, jobs with complementary resource requirements can be performed concurrently. For example, one initiator/terminator is assigned to a class requiring little CPU time and a great deal of I/O.

Multiprocessing Multiprocessing is a technique whereby the work of processing data is shared among two or more interconnected central processing units. CPU-to-CPU COMMUNICATION Some ways of communicating one CPU to another: At one extreme, communication may be represented by a few control signal lines that are userd broadly synchronize the operation of the CPU with that of another. By using a channel-to-channel adapter. ADVANTAGES OF MULTIPROCESSING •Increased availability •Increased production capacity •More efficient use of resources •Data sharing •Online problem solving •Inquiry and transaction processing

Related Documents