Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Distributed shared memory shyam soni


Published on

this presentation is created by Shyam soni.
if u like it then mail me

Published in: Education
  • Be the first to comment

Distributed shared memory shyam soni

  1. 1. DISTRIBUTED SHARED MEMORY 1 Shyam Soni 9501852909
  2. 2. INTRODUCTION ● DSM is a term in computer science that refers to a wide class of software and hardware implementations, in which each node of a cluster has access to shared memory in addition to each node's non-shared private memory. ●Distributed Shared Memory (DSM), also known as a distributed global address space (DGAS). ●The processors communicate with one another through various communication networks,such as high-speed buses or telephone lines. ●A distributed system is a collection of loosely coupled processors interconnected by a communication network.
  3. 3. Advantage There are four major advantage of distributed system 1. Resource Sharing 2. Computation Speedup 3. Reliability 4. Communication
  4. 4. Distributed shared memory
  5. 5. General Architecture of Distributed Shared Memory (DSM) System : ●It consists of number of nodes or machines or computer each of which is connected to each other through high speed communication channel. Each node has several components associated with it. ●A software memory mapping manager of each node maps its local memory into the memory mapping into the virtual shared memory. We call it the Distributed Shared Memory
  6. 6. Algorithims of DSM The algorithims for implementing DSM deal with two basic problems ➔ Static and Dynamic distribution of shared data across the system, to minimize access latency, and ➔Preserving a coherent view of shared data ,while minimizing coherence-management overhead.
  7. 7. Classifications of DSM Algorithims ●Single reader/Single writer algorithims ●Multiple reader/Single writer algorithims ●Multiple reader/Multiple writer algorithims
  8. 8. Design and Implementation Issues Thee are various factors that have to be kept In mind while designing and implementing the DSM systems. They are as follows: 1.Block Size: As we know, transfer of the memory blocks is the major operation in the DSM systems. Therefore block size matters a lot here. Size of the block depends on various factors like, paging overhead, thrashing, false sharing, and directory size.
  9. 9. 3.Replacement Strategy: It may happen that one node might be accessing for a memory block from DSM when its own local memory is completely full. In such a case, when the memory block migrating from remote node reaches, it finds no space to get placed. Thus a replacement strategy of major concern in the design and implementation of the DSM systems. Certain block must be removed so as to place the new blocks in such a situation. Several techniques are used for the replacement of old blocks such as removal of Least Recently Used memory blocks. 2.Structure of Shared Memory Space How the shared memory space is organized with data determines the structure of the shared memory space. It refers to the layout of shared data. It depends upon the type of application the DSM is going to handle.
  10. 10. Thank youThank you