Title: Design%20Issues%20of%20Prefetching%20Strategies%20for%20Heterogeneous%20Software%20DSM
1Design Issues of Prefetching Strategies for
Heterogeneous Software DSM
- Author Ssu-Hsuan Lu, Chien-Lung Chou,
Kuang-Jui Wang, Hsiao-Hsi Wang, and
Kuan-Ching Li - Speaker Chien-Lung Chou
- Date 2006/05/18
2Outline
- Introduction
- Motivation
- Related Work
- Proposed Method
- Performance Evaluation
- Conclusions and Future Work
3Introduction
- In Distributed Shared Memory (DSM) systems, it
induces - large number of page faults.
- large number of communication time.
4Introduction (cont.)
- Page faults and communication time are major
overheads in DSM systems. - We need additional strategies to reduce page
faults and communication time are - Home migration
- Write vector
- Prefetching
5Introduction (cont.)
- Most traditional prefetching strategies can
provide good performance in homogeneous cluster
platforms. - However, the performance of such strategies may
be worse in heterogeneous environment.
6Motivation
- Homogeneous cluster platform.
7Motivation (cont.)
- Heterogeneous cluster platform
8Motivation (cont.)
- We need to concern about heterogeneous
environment. - More and more personal computers will be perform
computations collectively. - Large number of advanced techniques will be
develop in this environment. - We will usually meet this environment in future.
9Related Work
- History Prefetching Strategy.
- It permits home nodes sending data to remote
nodes in advance. - It has some disadvantages
- Accumulated Waiting Phenomenon.
- Waiting Synchronization Phenomenon.
- Misprefetch.
- Home nodes have too much work.
10Related Work (cont.)
- Effective Prefetch Strategy
- Filtering Unnecessary Prefetches.
- Distributing Prefetch Overhead.
- Load Balancing with Barrier Synchronization.
- Agent Home of prefetching strategy
- It will find a node that will help home nodes to
transfer prefetching data. - Thus, it reduces overhead of home nodes.
11Proposed Method
Prefetching Strategy in Heterogeneous
Environments
12Proposed Method (cont.)
- According to above disadvantages, we propose the
method that allows - home nodes are adjusted to suitable place.
- high speed processors to execute prefetch in
advance. - low speed processors to leave the barrier early.
13Proposed Method (cont.)
- First, we distribute home pages to nodes that
have better resources. - These nodes are suitable to be home nodes because
they have better performance.
14Proposed Method (cont.)
- Second, we observe that hosts that have worse
resources finish work later, so we adjust policy
of prefetching strategy. - Originally, all hosts leave barrier at the same
time.
15Proposed Method (cont.)
- In our method, the hosts that have worse
resources leave barrier after requesting
prefetching data. - The hosts that have better resources leave
barrier after sending prefetching pages to hosts
that have worse resources.
16Proposed Method (cont.)
- Third, we also observe that hosts that have
better resources spend large amount of idle time
during barrier in heterogeneous environment. - It raises execution time and barrier time.
17Proposed Method (cont.)
- We utilize idle time in barrier of hosts that
have better resources to perform prefetching to
each other.
18Proposed Method (cont.)
19Performance Evaluation
- Experimental Platform - Hardware
Node ID CPU Memory Network
1 Intel P4 2.4GHz 256MB Fast Ethernet
2 Intel P4 1.7GHz 256MB Fast Ethernet
3 Intel P3 500MHz 640MB Fast Ethernet
4 Intel P2 350MHz 128MB Fast Ethernet
20Performance Evaluation (cont.)
- Experimental Platform - Software
- Linux Fedora Core 3.
- Kernel 2.6.9.
- JIAJIA DSM software.
21Performance Evaluation (cont.)
Host 1 Host 2 Host 3 Host 4
Idle Time in Barrier JIAJIA 70.71 67.79 27.00 0.36
Idle Time in Barrier History Prefetch 72.72 71.05 30.45 0.78
Idle Time in Barrier Effective Prefetch 68.31 67.42 26.60 0.63
Idle Time in Barrier Agent Home 70.72 68.61 28.11 0.63
Idle Time in Barrier Our Method 62.14 62.16 26.77 0.56
The Idle Time in Barrier for IS Application
22Performance Evaluation (cont.)
Host 1 Host 2 Host 3 Host 4
Idle Time in Barrier JIAJIA 3.75 7.35 8.67 8.21
Idle Time in Barrier History Prefetch 3.83 6.25 7.49 7.02
Idle Time in Barrier Effective Prefetch 3.92 6.29 7.43 6.97
Idle Time in Barrier Agent Home 4.34 6.03 7.03 6.57
Idle Time in Barrier Our Method 2.37 2.38 4.91 4.35
The Idle Time in Barrier for Merge
Application
23Performance Evaluation (cont.)
24Conclusions and Future Work
- In heterogeneous environment, benefits of
original prefetching strategies are limited. - In this paper, we utilized idle time to improve
overall performance. - In the best situation, our proposed method could
reduce idle time in barrier of about 60.
25Conclusions and Future Work (cont.)
- In the future, we will make effort to find a
method to optimize the use of idle time in
barrier. - In addition, we will also investigate the
parallel program execution with issues about
dynamic CPU loads include in our next development
stage.
26