Presentation is loading. Please wait.

Presentation is loading. Please wait.

November 1, 2005Sebastian Niezgoda TreadMarks Sebastian Niezgoda.

Similar presentations


Presentation on theme: "November 1, 2005Sebastian Niezgoda TreadMarks Sebastian Niezgoda."— Presentation transcript:

1 November 1, 2005Sebastian Niezgoda TreadMarks Sebastian Niezgoda

2 What is it? Distributed Shared Memory system Rice University project

3 DSM Many nodes in a cluster Each has limited private memory A shared memory available (large)

4 TreadMarks DSM Supports parallel computing Provides a shared address space for all machines in a cluster Globally shared memory but execution in private memory

5 TreadMarks DSM

6 Why use DSM? DSM provides an interface to the globally shared memory Every processor can access every data item without the programmer needing to know/worry how to do it How does MPI accomplish the “shared memory” access?

7 TreadMarks memory model Memory model defines how updates to shared memory are reflected to the processes in the system TreadMarks uses release consistency

8 TreadMarks release consistency Synchronization is used to prevent race conditions between for parallel processes Variable var is replicated to all processors Only next processor that gets the lock can access var and only that processor knows about the change to var  Less message traffic

9 TreadMarks release consistency Modification info is piggybacked to the lock grant message Invalidate protocol used  When acquired, modified pages are invalidated  Later access causes a miss and pages are re-acquired

10 TreadMarks multiple-writer Many processes can write to a page at the same time Page is copied when written to processor’s private memory Word-by-word comparison Diffs are used to update the original  Benefits?

11 TreadMarks Diff Creation

12 TreadMarks API Provides process creation, destruction, synchronization, shared memory allocation facilities A user-level library in Unix – no kernel mods needed (comm and mem mgmt already provided) Berkeley sockets for messaging  Every message is either a request or response

13 Performance Min round-trip time for smallest message 500 microseconds  Send 80 microseconds  Receive 80 microseconds  180 microseconds for wire time, interrupts

14 Performance cont. Time to remotely acquire free lock 827 microseconds 1149 microseconds if the manager did not have control of the block last 2792 microseconds to obtain a 4096 byte page from another processor (a remote access miss)

15 Performance cont. Time to make twin: 167 microseconds (4kb pages) Diff creation  430 microseconds if page is unchanged  472 microseconds if page changed  686 microseconds worst case (every other word is changed)

16 Applications Mixed Integer Programming Genetic Linkage Analysis

17 References “TreadMarks: Shared Memory Computing on Networks of Workstations” by Amza, Cox, Dwarkadas, Keleher, Lu, Rajamony, Yu, and Zwaenepoel. Rice University, Dept. of Computer Science http://www.cs.rice.edu/~willy/TreadMarks/ov erview.html http://www.cs.rice.edu/~willy/TreadMarks/ov erview.html http://en.wikipedia.org/wiki/Distributed_shar ed_memory


Download ppt "November 1, 2005Sebastian Niezgoda TreadMarks Sebastian Niezgoda."

Similar presentations


Ads by Google