Mentor Embedded Sourcery Analyzer D a t a s h e e t product features: ■■ Dynamic system analysis views Easily create, modify, and reorganize views in order to form and test theories about the data ■■ Large log file capability Proven capacity to handle multigigabyte datasets ■■ Capture system level behavior View the activity of processes and tasks, resources, and cores over time Sourcery Analyzer gives you insight into the performance of your system; it shows overall system activity, what’s running on each core, and activity of each process and thread. Navigate the vast sea of data to optimize performance Mentor® Embedded Sourcery™ Analyzer enables you to see the full picture of the performance of your embedded system. Modern embedded systems can produce vast volumes of performance, trace, log, and monitoring data. Traditional analysis methods require hand-crafted analysis scripts to sift through this data and provide rudimentary analysis. Sourcery Analyzer helps automate the process of collection and analysis, aiding comprehension of this data. Sourcery Analyzer offers unique and customizable visualization tools that enable the creation of customized performance dashboards. You can use the tool’s data calculation and reporting capabilities to efficiently extract metrics that are of interest to the individual developer. Work with multi-format, multi-OS trace data Sourcery Analyzer collects trace and profile data from a variety of facilities and plots system activity on a timeline. Sourcery Analyzer is OS-agnostic and can work with data from bare-metal, RTOS, or Linux® systems. Sourcery Analyzer also supports a vast variety of data formats ranging from text-based Event File Format (EFF), Linux Trace Tool, next generation (LTTng), and comma separated values (CSV). A flexible, Java script-based custom parsing engine is available for use with proprietary data formats. For Linux developers, Sourcery Analyzer integrates three sources of data. For system-level behavior, the popular and efficient LTTng tracing system is used to collect system level information. For application-level tracing, Mentor Embedded Sourcery™ CodeBench compilers can be used to automatically instrument function entries and exits. Lastly, Sourcery Analyzer users are provided with a simple facility for manually instrumenting relevant trace points into their application. w w w. m e nto r. co m /e m b e d d e d ■■ Resource utilization Understand how memory is consumed, interrupts responded to, and how applications and peripherals interact ■■ Unique computational abilities A built-in “Waveform Calculator” allows developers to perform mathematical, statistical, and logical operations on trace data ■■ Multi-OS, multi-format data support Works with CSV, text, and LTTng data formats from systems running bare-metal, RTOS, or Linux ■■ ‘Jump Start Program’ for successful adoption Product experts help with training, deployment, and customization benefits: ■■ Quickly discover and address bottlenecks With the cause of bottlenecks quickly identified, users are able to spend more time on real performance problems, not imagined ones ■■ Build more reliable multicore systems Hunt down the non-deterministic race conditions and deadlocks that can only be found by viewing events over time

Analyze then optimize Sourcery Analyzer helps institute a formalized method to analyze the performance of your system. Whether targeting high performance or low power, to improve your system you must first determine where to invest your time. Sourcery Analyzer allows you to see exactly where your system and its applications are spending time. It shows you the state of the cores in your system, the processes or tasks, and the resources. You can leverage this holistic view to identify and decide which portions of the system Sourcery Analyzer not only presents this holistic view, but also provides...