Resource Management Using Dynamical Load
Prediction and Multiprocessor Cooperation
ABSTRACT:
Resource management is a complicated problem in multiprocessor system. When tasks with real-time characteristic are scheduled on processor resource constraints such as CPU and memory have to be met, however, it is usually difficult for a system to ensure load balancing and resource constraints simultaneity. In order to handle this problem lots of algorithms have been brought forward. In this paper, a load balancing policy is proposed to manage and integrate processing resource of multiprocessor system which is used to send packets of packets flows on time. The policy dynamically predicts the wait time before a packet to be processed based on the load of the processor and the timestamp of the packet, and decides the migration of packet between processors to guarantee that the packets can be processed before deadline. This policy can not only increase the multiprocessor utilization but also enhance the capacity of the system against the jitter of load by packets migration between the processors as well. We use simulation experiment to show that the policy can reduce the possibility of packets delay and increase the concurrency of flows that the system can serve.
Existing System:
- In the existing system the proxies has been maintained in the critical path for each object updation or each proxy should connected with the centralized server.
- The consistency was not maintained while sharing the object.
- If the proxy has failed means the object has been lost.
- The existing system supports only single-object operations, and provides weak consistency semantics.
Disadvantages:
- Consistency was not maintained while migrating the object between the proxies.
- It does not handle the proxy disconnections.
- It supports only the single object operations.
Proposed System:
- This system forms the proxies in the tree structure. It shares the objects within the proxies. It reduces the workload in the server.
- Quiver enables consistent multiobject operations and optimizations for single-object reads that are not possible in these prior algorithms.
- This system recovers the proxy disconnection. The disconnected proxies maintained by alternate proxies or it will be maintained through server.
- This System use the kruskal’s algorithm for maintaining tree structure. It reduces weightage in the tree structure.
- It holds the object even when the proxy has been disconnected.
System Requirements
Hardware:
PROCESSOR : PENTIUM IV 2.6 GHz
RAM : 512 MB DD RAM
MONITOR : 15” COLOR
HARD DISK : 20 GB
FLOPPY DRIVE : 1.44 MB
CDDRIVE : LG 52X
KEYBOARD : STANDARD 102 KEYS
MOUSE : 3 BUTTONS
Software:
Front End : Java, Swing
Back End : MS Access
Tools Used : JFrameBuilder
Operating System : WindowsXP
Modules
· Create the centralized server and proxies.
· Object Migration from centralized server to proxies.
· Parent Proxy and Child Proxy Maintenance while disconnection.
· Proxy Tree Maintenance using Kruskal’s algorithm.
· Multiple object and Consistent object sharing.
Module Description
Module-1:
In this module we going to launch a centralized server , unbounded number of parent proxies and minimum number of child proxies. This server will provide the application to all the parent proxies.
Module-2:
This module deals with object migration its nothing but transferring our application from centralized server to the parent proxies. The server converts the whole application to an object. Then the parent proxies provides object to the child proxies.
Module-3:
This module provides the alternate solution when the proxies getting down. The client will get the response from some other proxies.
Module-4:
Here we are going to maintain the depth of the tree using kruskal’s algorithm. This algorithm calculates the round trip delay between the proxies .
Through that it will found the data round trip delay.
Module-5:
In this module we going to share the object consistently with out any interruption.
REFERENCE:
Xiuyan Guo, Wu Zhang, Jinlin Wang and Gang Wu, “Resource Management Using Dynamical Load Prediction and Multiprocessor Cooperation”, IEEE Conference, 2011.
