A Pseudo-Code for the Synchronization of Wait-and-Signal Tasks - Coursework Example

Systems Software PART A Q1: A Pseudo-code for the synchronization of wait-and-signal tasks: pWait [ ] : Get a dual semaphore token : } : pSignal [ ] : Release the dual semaphore token :} In this code, the dual semaphore is unavailable at first with a zero value. pWait is a high priority task running fast and makes the request to get the semaphore but due to the unavailability of the semaphore, it is blocked. The lower priority is then given to pSignal that runs and at some point releases the dual semaphore that unblocks the pWait task. Q3: The processor facilities needed by the nucleus need in order to perform its functions are: Interrupt Mechanism Protection of the Memory A set of Instructions Real-Time Clock The processor usually works faster than the peripherals it controls therefore it switches and works out other tasks as the peripheral complete whatever task they were doing. To allow for this overlap, the Input/Output peripherals should interrupt the processor when a peripheral is completing it’s task. This functionality is provided by the interrupt mechanism. The Interrupt Handler finds the origin of the interrupt and responds to it in the necessary manner. It usually carries out the tasks requested by the hardware releasing the interrupt. Q14: Four objectives of Memory Management; a) Relocation: In Multiprogrammed Computers, there is a memory that must be shared between the processors. It is not usually known in advance which programmes are in memory. Therefore, the absolute memories are usually out. In one process in finished, its free space is allocate to another process. The addresses for the processes in the system are transformed to physical addresses. b) Protection: Care must be taken so that one process does not access the address of other processes in the system. c) Logical Organization: This entails segmentation of address space on the system, which allows for cross referencing on one segment unto another by the system at the time the process is running. This gives protection to each segment making it read-only on execution. There segment sharing among the processes is well organized. d) Physical Organization: There is a two-level storage that should allow for the efficient flow of information from the Fast Main Memory to the Slow Secondary Memory. Q19: “Fragmentation” and ‘Compaction’: Fragmentation is a process that occurs on hard disk or other media of data storage where data is not written close to the physical space on the drive. For example, if the hard disk has to be read in piece of stored information from multiple storage places on the disk, it could not be easy to access the data if it had been stored or written on the same place on the drive. The process of fragmentation therefore involves separating data and organizing it on the disk drive. Compaction on the other hand refers to process of changing the portions of the memory so that the free portions can be aligned in a single big lock. Q20: Three replacement policies used in paged systems: 1. Clock Algorithm 2. FIFO Page Replacement 3. Recently Used The Recently used Policy is used as a predictor of the future. A page that has not been accessed for a very long time is discarded. This policy is tough to accurately implement and therefore the list of pages should be kept in a sorted manner with respect to reference time and the page should be marked with reference time. In the Fisrt-In-First-Out (FIFO) policy, the memory swaps the oldest page out with no regard to how recently it was created or accessed. Q22: IORB: Stands for Input/output Record Block Q26: How the I/O procedures cope with buffered applications: The I/O had uses buffer objects to lock-down the data structures used by user processes. The buffer objects refer to specific process units that are related to a process’s address range. When such an object is created, all the physical pages in its address range are locked in the memory and can be mapped twice in a system space. The locked pages cannot be made free until the deletion of the buffer objects. The I/O procedures utilize this feature by blocking the buffer object itself during the I/O process. The buffer object is prevented and unlocked during the CLEAN-UP process by the I/O procedure. After the protection of the buffer object, it can now be paged and swapped as a normal memory with access to its pages available in the buffer object. Q29: 'Block Linkage' method: Here, each file is linked to a group of disk blocks which may be scattered everyplace on the disk. A pointer is contained in the directory pointing to the first and the last block in the group of files. For instance, a file with five blocks might begin at block number 9 and end at bloc number 16, then block number 1, then block number 10 and finally block number 25 as shown in the diagram below: Q32: Secondary memory is the cheapest and lowest form of storing data and the secondary storage organization refers to how information and data is stored in a secondary storage device. File size affects secondary memory organization in different ways because it determines the number of files to be stored in certain storage space. For instance, block or pages with between 128-512 words is the unit size of the medium used for the storage of files. The filling system will then assign the required number of blocks to each file depending on the file size and the availability of storage space. Q33: Methods of dealing with free disk space: a) Bit Vector: Each block is represented by a single bit and a free block is represented by a 0. In the free-space bit map below, 15 blocks are free and the rest have been allocated. 110000110000001110011111100011111... b) Linked List: Here all the free disk blocks are linked together, while the pointer is located on the first free block. The first free block will also have a pointer pointing to the next free block and so on. c) Grouping: This is a modification of the free-list approach that is used to store the addresses of the number of free blocks in the free block appearing first. The first total-1 of these blocks are really free. The last block is the address of another block with all addresses of the total number of free blocks. Read More