Mahesh Kumar | Visvesvaraya Technological University (original) (raw)

Papers by Mahesh Kumar

One of the main goal of the operating system for the disk drives is to use the hardware efficient... more One of the main goal of the operating system for the disk drives is to use the hardware efficiently. we can meet this goal using fast access time and large disk bandwidth that depends on the relative positions of the read-write head and the requested data. Since memory management allows multiprogramming so that operating system keeps several read/write request in the memory. In order to service these requests, hardware (disk drive and controller) must be used efficiently. To support this in disk drive, the hardware must be available to service the request. if the hardware is busy, we can’t service the request immediately and the new request will be placed in the queue of pending requests. Several disk scheduling algorithms are available to service the pending requests. among these disk scheduling algorithms, the algorithm that yields less number of head movement will remain has an efficient algorithm.

In this research paper, we propose a new disk scheduling algorithm that will reduce the number of movement of head thereby reducing the seek time and it improves the disk bandwidth for modern storage devices. Our results and calculations show that, proposed disk scheduling algorithm will improve the performance of disk i/o by reducing average seek time compared to the existing disk scheduling algorithm. For few requests, the seek time and the total number of head movement is equal to SSTF or LOOK scheduling.

Most of sensor networks are static in nature and do not provide standardized and systematic netwo... more Most of sensor networks are static in nature and
do not provide standardized and systematic network
management capabilities. Intelligence is not associated with
the sensors. In order to make a sensor network more dynamic
in nature, there is need to monitor the sensor network with
the elements of the network management like Fault,
Configuration, Security and Performance. This project is on
the demonstration of Plug and Play (PnP) enabled devices
(smart sensor node) in different levels to provide network and
sensor management in dynamic fashion using Simple
Network Management Protocol (SNMP) and Hyper Text
Transfer Protocol (HTTP).

In nonpreemptive scheduling discipline, once the system has assigned a processor to a process, th... more In nonpreemptive scheduling discipline, once the system has assigned a processor to a process, the system cannot remove that processor from that process until it runs to completion or it voluntarily relinquishes its processor. In particularly, SJF scheduling algorithm which is nonpreemptive scheduling, selects the process with least in the length of the CPU burst time and the processor executes the process. The real difficulty with Shortest Job First (SJF) scheduling is that it requires the precise knowledge of how long a process will run and the length of the next CPU burst time and usually this information is not available. In this research paper, we proposed a dual simplex method to predict the length of the next CPU burst in SJF scheduling algorithm. Our calculations and results show that, the approximate/predict value of the next CPU burst will be similar or approximate to the length of the previous requests.

One of the main goals of the memory management is to allow multiprogramming. Several strategies a... more One of the main goals of the memory management is to allow multiprogramming. Several strategies are used to allocate a
memory to the process needed. These strategies require the entire process should be in the memory before execution and some of these strategies suffer from fragmentation. Virtual memory does not require entire process to be in memory before execution. It loads only those pages of the process in to the memory that are needed for execution. This can be achieved using paging scheme having number of frames in the memory to accommodate for each pages. Whenever one of the page in the memory need to be replaced, one of the page replacement algorithms are used. The performance of these page replacement algorithms depends on the total number of page faults achieved. In this paper, a new algorithm called Comparison Counting FIFO (CC-FIFO) and Distribution Counting FIFO (DC-FIFO) has been proposed using input enhancement technique. Our results and calculations show that the proposed algorithm minimizes the page fault rate compared to FIFO, LRU and Optimal page replacement algorithms.

Scheduling is a fundamental operating system function. Almost all computer resources are schedule... more Scheduling is a fundamental operating system function. Almost all computer resources are scheduled before use. All major CPU
scheduling algorithms concentrates more on to maximize CPU utilization, throughput, waiting time and turnaround time. In
particularly, the performance of round robin algorithm depends heavily on the size of the time quantum. To improve the performance of CPU and to minimize the overhead on the CPU, time quantum should be large with respect to the context switch time. Otherwise, context switching will be more. In this research paper, we propose a method to minimize the context switching and to break the fixed time quantum size in round robin scheduling algorithm using optimization techniques. Both results and calculations show that, our proposed method is more efficient than the existing round robin scheduling algorithm.