Virtutech presents Simics 4.0, the first deterministic multithreading virtualized software development platform.
Simics 4.0 Overview
This new version offers significant improvements in simulation performance and scalability, offers new user interface enhancements, and improvements in simulation modeling.
Performance and Scalability Improvements
Introducing Simics Accelerator(TM)
With the new Simics Accelerator(TM) set of technologies, Simics provides a unique set of scalability and execution speed capabilities. With Accelerator, Simics takes advantage of multiprocessor and multicore hosts to accelerate the execution of large target systems. Simics Accelerator also contains technology to exploit the properties of the target system in order to further improve scalability. The Simics infrastructure has also been improved to enhance parallelism, providing addition scalability benefits. The net effect of Simics Accelerator is a dramatic increase in speed on large target configurations compared to Simics 3.2. Speed increases of up to ten times over Simics 3.2 have been observed in preliminary testing on an eight-core host, running a 64-node network.
The effect on any particular target system will vary with the capacity of the host machine, as well as the properties of the target system.
User Interface Enhancements
Wind River Workbench ® Integration. Simics integrates into Wind River Workbench versions 3.0 or later, as an Eclipse plug in. Simics can be directly launched from with
Workbench, and target systems debugged both when stopped and running via a wdb connection.
Eclipse Integration. Virtutech provides a set of Eclipse plugins to start and control Simics sessions. The plugins work in a standard Eclipse installation, and this makes it easy to integrate Simics into Eclipse-based custom tool sets or workflows. The Simics plugins install directly from inside Eclipse, using the Eclipse package and update system.
Telnet control. In addition to the existing direct command-line interface, Simics can now be controlled across a telnet connection directly to Simics. This simplifies scripting and integration of Simics into existing enterprise simulation setups, and the use of Simics on processing servers and clusters.
Memory viewer. With the new memory content and memory system structure viewers, Simics makes it easy for a user to understand the target system setup and inspect its state.
Unicode support. Simics properly handles non-ASCII characters, including user data and filenames.
Modeling Improvements
Graphical performance meter. The graphical performance meter plots performance data live as Simics executes a target system. It provides tools to spot simulation performance issues and profile the simulator itself, helping users and modelers optimize their models for maximal virtualization performance.
 |
 |
Eclipse DML editor. DML editing is supported with a set of Eclipse plugins, providing keyword highlighting, automatic outlining, and interactive checking of your DML code. The DML editor provides interactive feedback on errors in your code, and fully encapsulates
the process of building DML models into an Eclipse environment.
Processor API. A new interface lets Simics users plug in existing instruction-set simulators into a Simics system. Detailed, cycle-accurate as well as fast performance-
optimized simulators are supported. The processors have full access to the rest of the system model in Simics, and can be combined with native Simics processor models and devices. Thus, the Simics infrastructure can be used for in-house experiments with computer architecture, leveraging existing processor models.
Improved DML debuggability. The C code generated by the DML compiler is now easier to debug and features better connections to the original DML source code.