Difference between revisions of "ENZO"

From Point
Jump to: navigation, search
Line 42: Line 42:
 
Here is a chart showing the contribution each function makes to the overall runtime. Notice that MPI communication time takes over 60% of the total runtime.  
 
Here is a chart showing the contribution each function makes to the overall runtime. Notice that MPI communication time takes over 60% of the total runtime.  
  
[[Image:http://giusto.nic.uoregon.edu/~scottb/enzo_files/MeanFunctionLinux.png]]
+
[[Image:MeanFunctionLinux.png]]
 +
 
 +
==Experiment Scalability==
 +
 
 +
Given the amount of time spend in MPI communications we do not expect this experiment to scale well. This chart shows that MPI communication time continues to dominate the runtime to an even greater extent at scale.
 +
 
 +
[[image:MeanRuntineAtScale.png]]
 +
 
 +
==Experiment Trace==
 +
This graphic shows how load imbalances cause long wait times for MPI_Allreduce. Some processors are experiencing as much as 8 seconds of wait time per reduce.
 +
 
 +
[[Image:trace.png|1000px]]
 +
 
 +
==Experiment Call-Paths==
 +
We observe the follow relationships in the experiment callpath:
 +
 
 +
* Almost all the time spend in MPI_Bcast is when it is called from MPI_Allreduce.
 +
* Almost all the time spend in MPI_Recv is when it is called from grid::CommunicationSendRegion.
 +
* Most all the time spend in MPI_Allgather is when it is called from CommunicationShareGrids.
 +
* Almost all the time spend in MPI_Allreduce is when it is called from CommunicationMinValue.
 +
 
 +
This chart show the details:
 +
 
 +
[[Image:CallpathRuntime3.png]]

Revision as of 21:09, 15 May 2008

ENZO Performance Study Summary

This is a short overview to the performance result of the ENZO application. For each experiment we used these inits/param files:

This is a relatively small experiment but was sufficent to generate some interesting performance results. For this study we used the TAU Performance System® to gather information about ENZO's performance, in particular we are interested in the Performance of the AMR simulation at scale. We ran these experiments on NCSA's Intel 64 Linux Cluster (Abe).

TAU Measurement overhead

Here is a short table listing the run-times for various experiments and the instrumentation overhead observed. Each run was on 64 processors (8 nodes).

Run Type Runtime (seconds) Overhead %
Uninstrumented runtime 1072 NA
Trace of only MPI event 1085 4.8%
Profile of all significant events 1136 6.0%
Profile with Call-path information 1196 11.6%
Profile of each Phase of execution 1208 12.7%

Runtime Breakdown on 64 processors

Here is a chart showing the contribution each function makes to the overall runtime. Notice that MPI communication time takes over 60% of the total runtime.

MeanFunctionLinux.png

Experiment Scalability

Given the amount of time spend in MPI communications we do not expect this experiment to scale well. This chart shows that MPI communication time continues to dominate the runtime to an even greater extent at scale.

MeanRuntineAtScale.png

Experiment Trace

This graphic shows how load imbalances cause long wait times for MPI_Allreduce. Some processors are experiencing as much as 8 seconds of wait time per reduce.

Trace.png

Experiment Call-Paths

We observe the follow relationships in the experiment callpath:

  • Almost all the time spend in MPI_Bcast is when it is called from MPI_Allreduce.
  • Almost all the time spend in MPI_Recv is when it is called from grid::CommunicationSendRegion.
  • Most all the time spend in MPI_Allgather is when it is called from CommunicationShareGrids.
  • Almost all the time spend in MPI_Allreduce is when it is called from CommunicationMinValue.

This chart show the details:

CallpathRuntime3.png