Using High-Rising Cities to Visualize Performance in Real-Time

Abstract

For developers concerned with a performance drop or improvement in their software, a profiler allows a developer to quickly search and identify bottlenecks and leaks that consume much execution time. Non real-time profilers analyze the history of already executed stack traces, while a real-time profiler outputs the results concurrently with the execution of software, so users can know the results instantaneously. However, a realtime profiler risks providing overly large and complex outputs, which is difficult for developers to quickly analyze. In this paper, we visualize the performance data from a real-time profiler. We visualize program execution as a three-dimensional (3D) city, representing the structure of the program as artifacts in a city (i.e., classes and packages expressed as buildings and districts) and their program executions expressed as the fluctuating height of artifacts. Through two case studies and using a prototype of our proposed visualization, we demonstrate how our visualization can easily identify performance issues such as a memory leak and compare performance changes between versions of a program. A demonstration of the interactive features of our prototype is available at https

Memory Cities: Visualizing Heap Memory Evolution Using the Software City Metaphor

“…Steinbrückner and Lewerentz [61], [62] adopted and extended this idea by visualizing the development history of software systems using elevated city maps. Software cities have been applied in the domains of concurrency visualization [22], software component communication and dependency visualization [11], [23]- [26], software performance visualization [63], [64], business process visualization [65], and test case analysis [66], [67]. Software cities have also been used in virtual reality [13], [14], [64], [68] and have been integrated into computer games such as Minecraft [69].…”

PADLA: A Dynamic Log Level Adapter Using Online Phase Detection

“…We employ a performance vector used in an existing performance profiling tool [12]. Each element in a vector represents the actual execution time consumed by a corresponding method in the interval.…”

“…Ogami et al proposed a three-dimensional cities visualization to show performance in real-time [12]. It visualizes the execution time for each method so that developers can recognize irregular behavior of a system.…”

PerfVis

“…Moreta and Telea [13] present a visualization of the behavior of the memory allocator in C programs to optimize functionality, decrease fragmentation, and improve response time. Another tool [14] facilitates the visualization of software performance in real-time, using the city metaphor to show the structure and performance of a program. The buildings in the city represent the classes in the system, and the heights of the buildings represent the number of times the methods of a class are called during execution.…”