• ETRI Journal
• /
• v.39 no.1
• /
• pp.124-134
• /
• 2017

Data races are one of the most difficult types of bugs in concurrent multithreaded systems. It requires significant time and cost to accurately detect bugs in complex large-scale programs. Although many race detection techniques have been proposed by various researchers, none of them are effective in all aspects. In this paper, we compare the performance of five recent dynamic race detection techniques: FastTrack, Acculock, Multilock-HB, SimpleLock+, and causally precedes (CP) detection. We experimentally demonstrate the strengths and weaknesses of these dynamic race detection techniques in terms of their detection capability, running time, and runtime overhead using 20 benchmark programs with different characteristics. The comparison results show that the detection capability of CP detection does not differ from that of FastTrack, and that SimpleLock+ generates the lowest overhead among the hybrid detection techniques (Acculock, SimpleLock+, and Multilock-HB) for all benchmark programs. SimpleLock+ is 1.2 times slower than FastTrack on average, but misses one true data race reported from Mutilock-HB on the large-scale benchmark programs.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.11
• /
• pp.1-7
• /
• 2017

In this paper, we introduce three monitoring filtering techniques which reduce the overheads of dynamic data race detection. It is well known that detecting data races dynamically in multi-threaded programs is quite hard and troublesome task, because the dynamic detection techniques need to monitor all execution of a multi-threaded program and to analyse every conflicting memory and thread operations in the program. Thus, the main drawback of the dynamic analysis for detecting data races is the heavy additional time and space overheads for running the program. For the practicality, we also empirically compare the efficiency of three monitoring filtering techniques. The results using OpenMP benchmarks show that the filtering techniques are practical for dynamic data race detection, since they reduce the average runtime overhead to under 10% of that of the pure detection.

• International Journal of Safety
• /
• v.1 no.1
• /
• pp.28-35
• /
• 2002

Kullback Discrimination Information (KDI) is one of the pattern recognition methods. KDI defined as a measure of the mutual dissimilarity computed between two time series was studied for detection and classification of bearing flaking on outer-race and inner-races. To model the damages, the bearings in normal condition, outer-race flaking condition and inner-races flaking condition were provided. The vibration sensor was attached by the bearing housing. This produced the total 25 pieces of data each condition, and we chose the standard data and measure of distance between standard and tested data. It is difficult to detect the flaking because similar pulses come out when balls pass the defection point. The detection and classification method for inner and outer races are defected by KDI and nearest neighbor classification rule is proposed and its high performance is also shown.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.01a
• /
• pp.7-8
• /
• 2019

자료경합은 두 개 이상의 스레드가 같은 공유메모리에 적절한 동기화 없이 접근하고, 적어도 한 개의 접근사건이 쓰기일 때 발생할 수 있는 동시성 오류이다. 자료경합은 프로그래머가 의도하지 않은 비결정적인 수행결과를 초래하여, 항공기 소프트웨어와 같은 고신뢰성이 요구되는 프로그램에서 치명적인 오류를 발생시켜 인적 물적 손해로 이어질 수 있다. 자료경합 탐지기법은 이러한 문제를 사전에 탐지하여 수정하는데 사용되어진다. 하지만 GPGPU 프로그램에서의 자료경합은 CPU 병행프로그램에서보다 복잡한 실행구조를 가지고 있어 스레드 및 메모리 계층, 스케줄링, 동기화 기법 등의 많은 변수가 존재한다. 이로 인해 실세계 프로그램에 자료경합 탐지기법을 적용하여 검증 시 이러한 변수들을 반영하여 실험하는데 많은 노력이 소요된다. 본 논문은 실세계 프로그램에서의 자료경합을 대표하는 4가지 패턴의 합성프로그램으로 이루어지고 실행 시 스레드 및 메모리 계층, 스레드 구조, 메모리 사용량 및 동기화 방안을 지정할 수 있는 벤치마크 모음을 제시한다.

• The KIPS Transactions:PartA
• /
• v.9A no.4
• /
• pp.525-534
• /
• 2002

Race conditions, races in short, need to be detected for debugging parallel programs, because the races result in unintended non-deterministic executions. To detect the races in an execution of program, previous techniques use a centralized data structure which may incur serious bottleneck in generating concurrency information, or show inefficient time complexity which depends on the degree of nested parallelism in comparing any two of them. We propose a new labeling scheme in this paper, which is scalable in generating the concurrency information without bottleneck by using private data structure, and improves time complexity into constant in checking concurrency. The scalability and time efficiency therfore makes on-the-fly race detection efficient not only for programs with either shared-memory or message-passing, but also for programs with mixed model of the two.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2016.07a
• /
• pp.1-2
• /
• 2016

다중 스레드 기반 병렬 프로그램에서의 자료경합 탐지는 동시에 수행되는 스레드 간의 비결정적인 상호작용 때문에 탐지하기 어려운 것으로 잘 알려져 있다. 동적 분석기술을 사용하여 자료경합을 탐지할 경우 프로그램 수행의 감시와 충돌하는 모든 메모리 연산의 분석을 위해 추가적인 오버헤드가 발생한다는 단점이 있다. 이러한 동적 분석의 추가적인 오버헤드를 줄이는 방법으로 감시 필터링 기술이 소개되고 있으며, 본 논문에서는 동적 자료경합 탐지를 위한 감시 필터링 기술 중 OpenMP 디렉티브 병렬 프로그램에 적용 가능한 두 기술을 대상으로 실용성과 효율성을 실험적으로 비교한다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.34 no.9
• /
• pp.395-406
• /
• 2007

Races in programs with OpenMP directives must be detected for debugging, because they may cause unexpected result by non-deterministic executions. But, Thread Checker of Intel corporation, a well-known existing tool for detecting the races, is not practical because this tool does not verify the existence of races and is known that the cost for race detection is too big. This paper presents a web-based tool which verify the existence of races with an optimal functionality and performance using the results from the property analysis of OpenMP program as well as the user requirements. Our tool is proved to be practical in the aspect of functionality and performance by experiments using synthetic programs, because the suggested tool can verify the existence of race and shows O(n) as the ratio of time consumption while Thread Checker can not verify the existence of race and shows $O(n^2)$ as the ratio, where n is the number of total accesses.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.4
• /
• pp.17-24
• /
• 2014

Detecting data races is important for debugging shared-memory programs with nested parallelism, because races result in unintended non-deterministic executions of the program. It is especially important to detect the first occurred data races for effective debugging, because the removal of such races may make other affected races disappear or appear. Previous dynamic detection tools for first race detecting can not guarantee that detected races are unaffected races. Also, the tools does not consider the nesting levels or need support of other techniques. This paper suggests a post-mortem tool which collects candidate accesses during program execution and then detects the first races to occur on the program after execution. This technique is efficient, because it guarantees that first races reported by analyzing a nesting level are the races that occur first at the level, and does not require more analyses to the higher nesting levels than the current level.

• Journal of KIISE:Software and Applications
• /
• v.31 no.7
• /
• pp.949-958
• /
• 2004

race detection using a boosting based algorithm requires a very large size of face and nonface data. In addition, the fact that there always occurs a need for adding additional training data for better detection rates demands an efficient incremental teaming algorithm. In the design of incremental teaming based classifiers, the final classifier should represent the characteristics of the entire training dataset. Conventional methods have a critical problem in combining intermediate classifiers that weight updates depend solely on the performance of individual dataset. In this paper, for the purpose of application to face detection, we present a new method to combine an intermediate classifier with previously acquired ones in an optimal manner. Our algorithm creates a validation set by incrementally adding sampled instances from each dataset to represent the entire training data. The weight of each classifier is determined based on its performance on the validation set. This approach guarantees that the resulting final classifier is teamed by the entire training dataset. Experimental results show that the classifier trained by the proposed algorithm performs better than by AdaBoost which operates in batch mode, as well as by ${Learn}^{++}$.

• Proceedings of the IEEK Conference
• /
• 2000.07b
• /
• pp.715-718
• /
• 2000

Although shared memory parallel programs are designed to be deterministic both in their final results and intermediate states, the races that occur when different processes access a common memory location in an order not guaranteed by synchronization could result in unintended non-deterministic executions of the program. So, Detecting races, particularly first data races, is important for debugging explicit shared memory parallel programs. It is possible that all data races reported by other on-the-fly algorithms would disappear once the first races were removed. To detect races parallel programs with nested loops and inter-thread coordination, it must guarantee the order of synchronization operations in an execution instance. In this paper, we propose an improved restructuring method that guarantee ordering execution instance and preserve the semantics of original program. This method requires O(np) time and (s + up) space, where n is the number of total operations, s is the number of synchronization operations and p is the number of parallelism in the execution. Also, this method makes on-the-fly detection of parallel program with nested loops and inter-thread coordination more easily in space and time complexity.