Memory leak detection in Plumbr

Abstract

SUMMARYPlatforms with automatic memory management, such as the JVM, are usually considered free of memory leaks. However, memory leaks can happen in such environments, as the garbage collector cannot free objects, which are not used by the application anymore, but are still referenced. Such unused objects can eventually fill up the heap and crash the application. Although this problem has been studied extensively, nevertheless, there are still many rooms for improvement in this area. This paper describes the statistical approach for memory leak detection, as an alternative, along with a commercial tool, Plumbr, which is based on the method. The tool is later analyzed with three case studies of real applications and in the process also analyzes strengths and weaknesses of the statistical approach for memory leak detection.

LeakSpot: detection and diagnosis of memory leaks in JavaScript applications

“…The memory leak problem has been addressed extensively in the context of other languages such as C++ and Java .…”

“…Leakbot works by analyzing the heap to find the suspicious data structures and then track the specific regions in the data structure for any growth. Plumbr uses a statistical approach for memory leak detection in Java programs. It uses statistical data about object age distribution to report the locations in the code where leaky objects are created.…”

“…While LeakSpot uses some of the techniques used in previous work for leak detection (such as staleness‐based technique of Bell and SWAT), those tools cannot be straightforward ported to JavaScript because they either modify Java Virtual Machine or instrument the binary code . LeakSpot profiles JavaScript application at the source level (instead of the Java Virtual Machine or binary level instrumentation).…”

“…A time‐count graph indicates a leak if it shows an increasing trend over time. Please note that according to these definitions of leaky collective‐staleness or time‐count graphs, we are considering the leaks that are increasing steadily over time . Therefore, these heuristics miss finding single‐leaked objects.…”

“…A memory leak in a managed language occurs when object references are unnecessarily maintained, resulting in the existence of unused objects on the heap. Because detecting leaked memory (unused objects) is an undecidable problem , many of the current solutions to find leaked memory use heuristics ; however, there are challenges and limitations in applying those tools and techniques on JavaScript applications. First, most of these tools are tuned for a specific environment, such as Java/C++ programs; they either instrument the Java Virtual Machine (JVM) or the binary code of the applications and, therefore, cannot be used for JavaScript applications running in the browser.…”