(1) For the CPU, there are many processes, which one needs to be done first, or which one needs to be done later? This is scheduling;
(2) job status: the job generally needs to go through four stages: "Submit", "backup", "run", and "complete". (3) submission status: after a job is submitted to the data center, the job is in the submitted status after being input through the spooling system process or when the user enters the job to the computer through a terminal. (4) backup status: the job has been input to the disk input well through the spooling system. Waiting for the job to be transferred to the memory for running, the job is in the backup status. To manage and schedule the job, set a job control block (JCB) for each job. The job control block records messages such as the control type and resource requirements. (5) allocate necessary resources after the job scheduler selects a parent job in the backup Job Queue. It is recommended that the job enter the running state after a group of corresponding processes are transferred to the memory; each process status (process running state, memory process ready state, memory blocking state, external process ready state, external process blocking state, etc.) corresponds to the job running state. (6) completion status: when the process ends normally or is terminated due to an error, the job enters the completion status; the termination process is responsible for handling the aftermath; others ------------------ (7) advanced (long-term) scheduling -- Job Scheduling
Advanced scheduling is also called Job Scheduling. Job Scheduling is used to determine which jobs on the external input well in the job reserve column are transferred to the memory, allocate necessary resources for them, and create processes, set the Process status to the ready state, and then rank the newly created process in the ready queue to compete for the CPU. In the batch processing system, after the job enters the system, it first resides in the input well of external storage, so job scheduling is required to load them into memory in batches. However, in a time-sharing system, in order to respond in a timely manner; the commands and data entered by the user through the keyboard are directly in the memory, so no job scheduling is required. Similar to real-time systems, Job Scheduling is usually not required;(8) low-level (short-term) scheduling-process scheduling determines which process in the ready queue will obtain the processor. Then, the allocation program executes the operation to allocate the processor to the process; scheduling runs frequently. In a time-sharing system, it usually runs once for dozens of milliseconds. process scheduling is the most basic scheduling, and any operating system has process scheduling; scheduling (9) preemptive
Mode) This scheduling method allows the process scheduler to terminate a running process according to a certain principle, assign the process to the processor, and re-assign the process to another process. The preemption principle is as follows: each process runs on a time slice. After a time slice is used up, the process is terminated and re-scheduled. This principle applies to the time-sharing system. The priority principle is as follows: it usually gives higher priority to some important and urgent processes. When such processes enter the ready queue, for example, converting from the blocking state to the ready state, or, when a static ready state is designed for an active ready state or a newly created process that enters the ready state queue, if the priority is higher than that of the process being executed, the processes being executed are prohibited; distribute the processor to a high-priority process to execute it; --------------------------------------------------------------- (10) scheduling mode and algorithm selection criteria and evaluation (Batch Processing System/time-sharing system) (11) job/process scheduling algorithm A: FCFS B: short job/process first --- scheduling algorithm; C: high response rate first(12) high response rate first: according to the principle of limited high response rate, calculate the Response Ratio of each job in the backup Job Queue to Rp, and then select the job with the highest value for running; RP = (Waiting Time + required running time)/required running time OR = 1 + waiting time/required running timeWaiting time = scheduling time-
Arrival time; turnover time = completion time
-Submission time;
