IDEAL Research

About

Sim-SODA (SOftware Dependability Analysis) is a unified simulation framework for estimating microprocessor reliability in the presence of soft errors. Originally designed for in-lab use, Sim-SODA is being released to the general public and can provide researchers with highly accurate estimates of the reliability of different hardware structures across different workloads in the form of the metric AVF (Architectural Vulnerability Factor). Built on top of the Sim-Alpha platform, Sim-SODA provides some substantial improvements to previous research in this area:

Unified approach for modeling hardware features
Offers high levels of granularity
Covers more hardware structures, in more detail
Open source and free license
Provides AVF estimations for hardware including:
- Instruction Queue
- Load/Store Queue
- Reorder Buffer
- Register File
- Data Level 1 Cache
- Translation Lookaside Buffer
- Victim Buffer
- Function Unit

Documentation

Xin Fu, Tao Li and Jose Fortes, "Sim-SODA: A Unified Framework for Architectural Level Software Reliability Analysis" Workshop on Modeling, Benchmarking and Simulation (Held in conjunction with International Symposium on Computer Architecture), June 2006

If you use Sim-SODA for your own published work, we kindly ask that you cite the above paper in your references.

Current Version

The current version of the sim-soda is version 1.1.0:

File Name: simsoda-1.1.0.tar.gz
File Size: 616991 Bytes
MD5: fb112d0f6716805d3213197b906416cb

Acknowledgements

This project was supported by the National Science Foundation under Grants No. CSR-0720476; NASA award no. NCC 2-1363; the Microsoft Research Trustworthy Computing award no. 14707.

Download

Sim-SODA is released freely under the GNU GPL and may be downloaded, executed, copied, and modified without limitation by the end user. No user may place any restrictions on the use of this software, including as modified by the user. Sim-SODA is provided as is, with no warranty of any kind, express or implied. The user of the program accepts full responsibility for the application of the program and the use of any results produced.

While registration is optional, we ask that you fill out as much of the information below as possible, so that we can have an estimate as to the number and sector of end users of Sim-SODA. No collected information will be used for any purpose other than statistical analysis of the Sim-SODA.

	Name:
	Email:
	University/Company:
	Use:
	Would you like to be notified when new versions are released:
Please enter the verification code above: