Taxim: A Toolchain for Automated and Configurable Simulation for Embedded Multiprocessor Design

January 13, 2016 Β· Entered Twilight Β· πŸ› arXiv.org

πŸŒ… TWILIGHT: Old Age
Predates the code-sharing era β€” a pioneer of its time

"Last commit was 10.0 years ago (β‰₯5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: Benchmarks, Experimenter.py, Experiments, Extractor.jar, Instructions, LICENSE, OutputMetrics, Paths, Power, Precheck.py, Results, Source, Topologies, TopologyCreator-v.2.4.jar, TopologyList, TopologyNamer.py

Authors Gorker Alp Malazgirt, Deniz Candas, Arda Yurdakul arXiv ID 1601.03341 Category cs.DC: Distributed Computing Citations 1 Venue arXiv.org Repository https://github.com/bouncaslab/TaXim/ ⭐ 2 Last Checked 2 months ago
Abstract
Multicore embedded systems have been constantly researched to improve the efficiency by changing certain metrics, such as processor, memory, cache hierarchies and their cache configurations. Using Multi2Sim and McPAT simulators in combination allows the user to design various multiprocessing architectures and estimate performance, power, area and timing metrics. However, the design time required to simulate these systems is daunting and prone to human error. In this paper, we introduce Taxim, a toolchain that can automatically create requested multicore on-chip topologies along with minimizing the simulation time due to repetitive tasks between architectural power, energy and timing simulations. Taxim's decision-tree-based topology synthesis tool creates processor configuration files that can be highly erroneous when generated manually. The toolchain also automates the steps from design entry to output report extraction by running automation scripts, and listing the results. Our experiments show that multiprocessing architectures with 32 cores and irregular cache hierarchies are more than 1k lines of code in Multi2Sim's processor configuration format and Taxim can create such a file in less than 10 milliseconds. The source code is freely available at https://github.com/bouncaslab/TaXim/.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

πŸ“œ Similar Papers

In the same crypt β€” Distributed Computing