|
| About DGSim |
 |
|
| why another simulator? |
|
|
|
The Delft Grid Simulator (DGSim) was born around August 2007 from the need to
simulate various grid resource management architectures. The goal of DGSim
is to simplify the simulation of multi-grid systems, that is, of systems that
comprise multiple grids.
|
|
|
|
The DGSim is built around the concept of experiment, that is,
a group of simulations that cover (provide insights and deeper knowledge about) a family of real-world phenomena.
A scenario is a group of simulations that cover a single phenomena.
Thus, an experiment comprises one or several scenarios, which in turn comprise
one or several related simulations.
|
|
|
|
DGSim focuses on three aspects: simulation of many grid resource management architectures,
on generating and replaying grid workloads, and on helping with the
simulation management (i.e., managing many repetitions of a simulation,
managing simulations with many parameter values, providing results processing tools,
and enabling collaborative experiments).
By comparison, its alternatives
(e.g., GridSim, SimGrid, GangSim, ChicSim, etc.) focus either on providing a low-level
simulation framework (e.g., GridSim, SimGrid), or on orthogonal grid resource
management issues (e.g., GangSim focuses on usage service level agreements).
Neither of these alternatives offers suitable simulation management solutions;
in particular, they have problems with the large-scale simulations required
by multi-grid environments.
|
|
|
| Download |
|
|
| try this simulator |
|
|
|
Python implementation using an SQLite database backend coming soon.
|
|
|
| People |
|
|
| who built this simulator? |
|
|
|
Alexandru Iosup. Contact person: A.Iosup@tudelft.nl.
|
|
|
|
Ozan Sonmez
|
|
|
|
Shanny Anoep
|
|
|
|
Dick Epema
|
|
|
| DGSim design |
|
|
| an overview of the DGSim design |
|
|
|
|
|
|
|
We envision four main roles for the DGSim community member. The
Experiment Designer creates experiment-specific configuration
files for the DGSim. The Simulator Developer develops simulation modules
for new scenarios. The Quality Assurance (QA) Team Member cetifies the
correctness of the results. The Investigator analyzes the experiment results.
|
|
|
|
The Experiment Manager toolbox controls the execution of the experiments.
|
|
|
|
The Workload Generation toolbox can load existing workloads or generate new workloads, based on a given model.
The workload model parameters may be specified by the Experiment Designer, extracted
automatically from existing traces, or iteraticely computed until the generated workload
has the characteristics specified by the Experiment Designer.
|
|
|
|
The Simulation toolbox can simulate the experiment scenarios, and
store the results in a structure that facilitates further processing.
|
|
|
|
The Data Warehousing toolbox processes the results of the experiments
into usable data for both generic and experiment-specific purposes.
|
|
|
| DGSim features |
|
|
| the main features of our simulator |
|
|
|
DGSim currently simulates six grid resource management architectures:
independent clusters operated with job push (sep-c),
independent clusters operated with matchmaking (condor),
centralized meta-scheduling with job pull (cern),
centralized meta-scheduling with job push (koala),
federated clusters with matchmaking (fcondor), and
inter-operated grids with delegated matchmaking (dmm).
|
|
|
|
DGSim can load workloads in the Grid Workload Format (GWF) of the Grid Workload Archive,
and in the Synthetic Workload Format (SWF) of the Parallel Workload Archive.
|
|
|
|
DGSim enables saves provenance data, which .
|
|
|
| Related work |
|
|
| other grid simulators that you might find useful |
|
|
|
GridSim
|
|
|
|
SimGrid
|
|
|
|
GangSim
|
|
|
|
ChicSim
|
|
|
| Research using DGSim |
|
|
| who uses this simulator? |
|
|
|
A. Iosup, D.H.J. Epema, T. Tannenbaum, M. Farrellee, M. Livny,
Inter-Operating Grids through Delegated MatchMaking
![Delegated MatchMaking Article, PDF [1.5MB]](icos/icopdf.gif) ,
In the ACM/IEEE SuperComputing Conference on High Performance Networking and Computing (SC'07),
Nov 10-16, 2007.
info used DGSim for comparing the performance and the overhead of various grid scheduling alternatives.
|
|
|
|
A. Iosup, M. Jan, O. Sonmez, and D.H.J. Epema, On the Dynamic Resource Availability in Grids,
In the IEEE Conference on (Grid2007), Sep 19-21, 2007.
info used DGSim for evaluating the impact of dynamic resource availability on the grid performance.
|
|
|
|
|
The Delft Grid Simulator (DGSim)
