Visual Solution to High Performance Computing

1
Visual Solutionto High
Performance Computing

• Computer and Automation Research Institute
• Laboratory of Parallel and Distributed
  Systemshttp//www.lpds.sztaki.hu/pgrade

2
The Big Issue
Parallelization?
Sequentialprogram code C, C Fortran
Parallelprogram code ?
Debugging?
Performance?
Execution?
3
Distributed supercomputing with P-GRADE

• Parallel Grid Run-time and
• Application Development Environment
• (P-Grade)
• The first highly integrated parallel application
  development system for Grid and clusters
• Provides
• Parallel, supercomputing programming
• Fast and efficient development
• Observation and visualization
• Fault and performance analysis
• ON CLUSTERS
• ON SUPERCOMPUTERS
• IN THE GRID

4
DESIGN Three hierarchical layers for easy
parallelisation
Application window
Process window
Text window
5
DESIGN Built-in Communication Templates for fast
design
6
DEBUGGING Hierarchical control and views to
eliminite programming errors
7
DEBUGGING Macrostep-by-macrostep

• Support for systematic debugging to handle
  non-deterministic behaviour of parallel
  applications
• Automatic dead-lock detection
• Replay technique with collective breakpoints
• Active control generation of Execution Trees
• Testing parallel programs by temporal logic
  assertions

8
MONITORING statistics and event traces for
performance evaluation

• Collection of both statistics and event traces
• Application monitoring and visualization in the
  Grid
• Evaluation of long-running programs based on
  semi-on-line trace collection
• Support for debugger in
• P-GRADE by execution visualisation

9
PERFORMANCE VISUALIZATION Visualization of Event
Traces

• User controlled focus on processors, processes
  and messages

10
P-GRADE Stages of Software Development and
Execution
Design, debugging
Performance- analyzis
Execution
Grid Cluster Supercomputer
11
Application I Ultra-short range weather forecast
(Hungarian Meteorology Service)
Forecasting dangerous weather situations (storms,
fog, etc.), crucial task in the protection of
life and property
Processed information surface level
measurements, high-altitude measurements, radar,
satellite, lightning, results of previous
computed models
Result

• Requirements
• Execution time lt 10 min
• Higher resolution

12
Processing of satellite images
Interpolation to basic GRID
Transformation
Processing Altitude of clouds
13
P-GRADE in meteorology demonstrated at Euro-Par
2003 conference

P-GRADE GUI
HungarianMeteorologicalService

DB
P-GRADE - RS CONDOR-PVM
Parallel execution and Monitoring
14
Workflow support (in the Grid)Illustrated using
the meteorological application
15
APPLICATION II Urban traffic simulation(Universi
ty of Westminster)
16
Checkpointing and migration of P-GRADE programs
between clusters
London-UoW
Budapest, MTA SZTAKI
SZTAKI BUTE
17
Monitoring visualisation - job migration in
the Grid -
Slow-down of simulator job due to overloaded
cluster
Migration of job between clusters in the GRID
18
Summary Advantages of P-GRADE

• Effective and efficient support for each stage
  of the parallel program development
• Fast parallelization of existing algorithms
• Reusability of sequential code
• Hiding of low level communication functions
• Integrated graphical concept
• Predefined communication templates
• Support for hierarchical design
• Even non-professional programmers can use
  it (without a steep learning curve)
• Portability (supercomputers, PC clusters,
  Grid)

19

• Parallel Grid Run-time and Application
  Development Environment