Linhai Song

Understanding and detecting real-world performance bugs

Developers frequently use inefficient code sequences that could be fixed by simple patches. These inefficient code sequences can cause significant performance degradation and resource waste, referred to as performance bugs. Meager increases in single threaded performance in the multi-core era and increasing emphasis on energy efficiency call for more effort in tackling performance bugs. This paper conducts a comprehensive study of 110 real-world performance bugs that are randomly sampled from five representative software suites (Apache, Chrome, GCC, Mozilla, and MySQL). The findings of this study provide guidance for future work to avoid, expose, detect, and fix performance bugs. Guided by our characteristics study, efficiency rules are extracted from 25 patches and are used to detect performance bugs. 332 previously unknown performance problems are found in the latest versions of MySQL, Apache, and Mozilla applications, including 219 performance problems found by applying rules across applications.

Automated atomicity-violation fixing

Fixing software bugs has always been an important and time-consuming process in software development. Fixing concurrency bugs has become especially critical in the multicore era. However, fixing concurrency bugs is challenging, in part due to non-deterministic failures and tricky parallel reasoning. Beyond correctly fixing the original problem in the software, a good patch should also avoid introducing new bugs, degrading performance unnecessarily, or damaging software readability. Existing tools cannot automate the whole fixing process and provide good-quality patches. We present AFix, a tool that automates the whole process of fixing one common type of concurrency bug: single-variable atomicity violations. AFix starts from the bug reports of existing bug-detection tools. It augments these with static analysis to construct a suitable patch for each bug report. It further tries to combine the patches of multiple bugs for better performance and code readability. Finally, AFixs run-time component provides testing customized for each patch. Our evaluation shows that patches automatically generated by AFix correctly eliminate six out of eight real-world bugs and significantly decrease the failure probability in the other two cases. AFix patches never introduce new bugs and usually have similar performance to manually-designed patches.

Toddler: detecting performance problems via similar memory-access patterns

Performance bugs are programming errors that create significant performance degradation. While developers often use automated oracles for detecting functional bugs, detecting performance bugs usually requires time-consuming, manual analysis of execution profiles. The human effort for performance analysis limits the number of performance tests analyzed and enables performance bugs to easily escape to production. Unfortunately, while profilers can successfully localize slow executing code, profilers cannot be effectively used as automated oracles. This paper presents Toddler, a novel automated oracle for performance bugs, which enables testing for performance bugs to use the well established and automated process of testing for functional bugs. Toddler reports code loops whose computation has repetitive and partially similar memory-access patterns across loop iterations. Such repetitive work is likely unnecessary and can be done faster. We implement Toddler for Java and evaluate it on 9 popular Java codebases. Our experiments with 11 previously known, real-world performance bugs show that Toddler finds these bugs with a higher accuracy than the standard Java profiler. Using Toddler, we also found 42 new bugs in six Java projects: Ant, Google Core Libraries, JUnit, Apache Collections, JDK, and JFreeChart. Based on our bug reports, developers so far fixed 10 bugs and confirmed 6 more as real bugs.

Statistical debugging for real-world performance problems

Design and implementation defects that lead to inefficient computation widely exist in software. These defects are difficult to avoid and discover. They lead to severe performance degradation and energy waste during production runs, and are becoming increasingly critical with the meager increase of single-core hardware performance and the increasing concerns about energy constraints. Effective tools that diagnose performance problems and point out the inefficiency root cause are sorely needed. The state of the art of performance diagnosis is preliminary. Profiling can identify the functions that consume the most computation resources, but can neither identify the ones that waste the most resources nor explain why. Performance-bug detectors can identify specific type of inefficient computation, but are not suited for diagnosing general performance problems. Effective failure diagnosis techniques, such as statistical debugging, have been proposed for functional bugs. However, whether they work for performance problems is still an open question. In this paper, we first conduct an empirical study to understand how performance problems are observed and reported by real-world users. Our study shows that statistical debugging is a natural fit for diagnosing performance problems, which are often observed through comparison-based approaches and reported together with both good and bad inputs. We then thoroughly investigate different design points in statistical debugging, including three different predicates and two different types of statistical models, to understand which design point works the best for performance diagnosis. Finally, we study how some unique nature of performance bugs allows sampling techniques to lower the overhead of run-time performance diagnosis without extending the diagnosis latency.

ECON: An Approach to Extract Content from Web News Page

This paper provides a simple but effective approach, named ECON, to fully-automatically extract content from Web news page. ECON uses a DOM tree to represent the Web news page and leverages the substantial features of the DOM tree. ECON finds a snippet-node by which a part of the content of news is wrapped firstly, then backtracks from the snippet-node until a summary-node is found, and the entire content of news is wrapped by the summary-node. During the process of backtracking, ECON removes noise. Experimental results showed that ECON can achieve high accuracy and fully satisfy the requirements for scalable extraction. Moreover, ECON can be applied to Web news page written in many popular languages such as Chinese, English, French, German, Italian, Japanese, Portuguese, Russian, Spanish, Arabic. ECON can be implemented much easily.

Fixing, preventing, and recovering from concurrency bugs

Concurrency bugs are becoming widespread with the emerging ubiquity of multicore processors and multithreaded software. They manifest during production runs and lead to severe losses. Many effective concurrency-bug detection tools have been built. However, the dependability of multi-threaded software does not improve until these bugs are handled statically or dynamically. This article discusses our recent progresses on fixing, preventing, and recovering from concurrency bugs.

What change history tells us about thread synchronization

Multi-threaded programs are pervasive, yet difficult to write. Missing proper synchronization leads to correctness bugs and over synchronization leads to performance problems. To improve the correctness and efficiency of multi-threaded software, we need a better understanding of synchronization challenges faced by real-world developers. This paper studies the code repositories of open-source multi-threaded software projects to obtain a broad and in- depth view of how developers handle synchronizations. We first examine how critical sections are changed when software evolves by checking over 250,000 revisions of four representative open-source software projects. The findings help us answer questions like how often synchronization is an afterthought for developers; whether it is difficult for devel- opers to decide critical section boundaries and lock variables; and what are real-world over-synchronization problems. We then conduct case studies to better understand (1) how critical sections are changed to solve performance prob- lems (i.e. over-synchronization issues) and (2) how soft- ware changes lead to synchronization-related correctness problems (i.e. concurrency bugs). This in-depth study shows that tool support is needed to help developers tackle over-synchronization problems; it also shows that concur- rency bug avoidance, detection, and testing can be improved through better awareness of code revision history.

Simultaneous Product Attribute Name and Value Extraction from Web Pages

Much work has been done in the area of template independent web data extraction. However, these approaches deal with the attribute value extraction and annotation either in separate phases or constrained to a predefined set of attributes which is highly ineffective. In this paper, we perform the attribute extraction and annotation simultaneously by extracting the attribute name and value pair at the same time. In our approach, we use a co-training algorithm with naive Bayesian classifier to identify the candidate attribute name and value pairs in the unlabeled pages. The candidate attribute name and value pairs are used to detect the specification block of the product in web pages. Finally, all the attribute name and value pairs in the specification block are discovered. We conduct experiments for three types of products and obtain a promising result.

ContentEx: A framework for automatic content extraction programs

Web pages are often decorated with extraneous information (such as navigation bars, branding banners, JavaScript and advertisements). This kind of information may distract users from actual content they are really interested in and may reduce effects of many advanced web applications. Automatic content extraction has many applications ranging from providing data for web mining to realizing better accessing the web over mobile devices. In this paper, we propose ContentEx, a framework for automatic content extraction programs, which we use to organize codes of automatic content extraction programs and to facilitate the development of related solutions. We also introduce how we extract content from forum pages in this framework to fulfill the requirement from our actual application.

COMP: Compiler Optimizations for Manycore Processors

Applications executing on multicore processors can now easily offload computations to many core processors, such as Intel Xeon Phi coprocessors. However, it requires high levels of expertise and effort to tune such offloaded applications to realize high-performance execution. Previous efforts have focused on optimizing the execution of offloaded computations on many core processors. However, we observe that the data transfer overhead between multicore and many core processors, and the limited device memories of many core processors often constrain the performance gains that are possible by offloading computations. In this paper, we present three source-to-source compiler optimizations that can significantly improve the performance of applications that offload computations to many core processors. The first optimization automatically transforms offloaded codes to enable data streaming, which overlaps data transfer between multicore and many core processors with computations on these processors to hide data transfer overhead. This optimization is also designed to minimize the memory usage on many core processors, while achieving the optimal performance. The second compiler optimization re-orders computations to regularize irregular memory accesses. It enables data streaming and factorization on many core processors, even when the memory access patterns in the original source codes are irregular. Finally, our new shared memory mechanism provides efficient support for transferring large pointer-based data structures between hosts and many core processors. Our evaluation shows that the proposed compiler optimizations benefit 9 out of 12 benchmarks. Compared with simply offloading the original parallel implementations of these benchmarks, we can achieve 1.16x-52.21x speedups.