In this paper, we present an approach to automatically detect high impact coding errors in large Java applications which use frameworks. These high impact errors cause serious performance degradation and outages in re...
详细信息
ISBN:
(纸本)1581138202
In this paper, we present an approach to automatically detect high impact coding errors in large Java applications which use frameworks. These high impact errors cause serious performance degradation and outages in real world production environments, are very time-consuming to detect, and potentially cost businesses thousands of dollars. Based on 3 years experience working with IBM customer production systems, we have identified over 400 high impact coding patterns, from which we have been able to distill a small set of pattern detection algorithms. These algorithms use deep static analysis, thus moving problem detection earlier in the development cycle from production to development. Additionally, we have developed an automatic false positive filtering mechanism based on domain specific knowledge to achieve a level of usability acceptable to IBM field engineers. Our approach also provides necessary contextual information around the sources of the problems to help in problem remediation. We outline how our approach to problem determination can be extended to multiple programming models and domains. We have implemented this problem determination approach in the SABER tool and have used it successfully to detect many serious code defects in several large commercial applications. This paper shows results from four such applications that had over 60 coding defects. Copyright 2004 ACM.
作者：
Papaharalabos, S.Sweeney, P.Evans, B.G.
University of Surrey Guildford Surrey GU2 7XH United Kingdom CCSR
Department of Electronic Engineering University of Surrey Guildford Surrey GU2 7XH United Kingdom
In this paper two normalization methods are considered to improve the performance of the iterative soft output Viterbi algorithm (SOVA). This is mainly because of the nature of the SOVA, which is a sub-optimum algorit...
详细信息
We proved that the hardware implementation of the proposed code and the new packet synchronization system was effectively realized by using a unique circuit configuration. A three-dimensional size-five coder and decod...
详细信息
ISBN:
(纸本)0780377281
We proved that the hardware implementation of the proposed code and the new packet synchronization system was effectively realized by using a unique circuit configuration. A three-dimensional size-five coder and decoding-synchronization system was implemented on FPGA. The developed FPGA was applied to a high-speed MPEG communication device, which can transmit a movie signal of 20 Mbps.
On error exponents in the simple quantum hypothesis testing problem, a noncommutative analogue of the Hoeffding bound [1] is presented. If the hypotheses, namely two density operators, are mutually commutative, the bo...
详细信息
ISBN:
(纸本)0780377281
On error exponents in the simple quantum hypothesis testing problem, a noncommutative analogue of the Hoeffding bound [1] is presented. If the hypotheses, namely two density operators, are mutually commutative, the bound is identical with the classical counter part and thus is optimal. As a by-product, a simple proof of the direct part of the quantum Stein's lemma [2] [3] is also given.
Peer-to-peer networks are networks of heterogeneous computers sharing files or services. This paper proposes to use a data storage scheme using maximum distance separable codes to optimize the dissemination of the dat...
详细信息
ISBN:
(纸本)0780377281
Peer-to-peer networks are networks of heterogeneous computers sharing files or services. This paper proposes to use a data storage scheme using maximum distance separable codes to optimize the dissemination of the data in the network in order to globally enhance the data access.
A framework for the analysis of error propagation for high rate runlength-limited codes is developed. The susceptibility to errors of capacity approaching code constructions, the enumerative and sequence replacement t...
详细信息
A framework for the analysis of error propagation for high rate runlength-limited codes is developed. The susceptibility to errors of capacity approaching code constructions, the enumerative and sequence replacement technique and bit stuffing, are analyzed in detail. It is shown that the first two techniques experience low error propagation and that they are superior to the popular bit-stuffing technique.
Distributed error correction in a network when a source message is transmitted to a set of receiving nodes on a network is addressed. Such an approach is called network error correction, with the conventional link-by-...
详细信息
Distributed error correction in a network when a source message is transmitted to a set of receiving nodes on a network is addressed. Such an approach is called network error correction, with the conventional link-by-link approach being a special case. Network generalizations of the Hamming bound and the Gilber-Varshamov bound are derived.
A class of codes recently introduced and now being used for digital system test compaction are addressed. Theorems and constructive bounds are given for certain classes of X-codes. It is shown that minimal codes are g...
详细信息
A class of codes recently introduced and now being used for digital system test compaction are addressed. Theorems and constructive bounds are given for certain classes of X-codes. It is shown that minimal codes are guaranteed to exist, but are not necessarily unique or non-trivial.
Objective To determine the accuracy of routine data coding in a large multispeciality urological unit. Materials and methods From the clinical records, the diagnosis and procedure codes were ascribed to 106 finished c...
详细信息
Objective To determine the accuracy of routine data coding in a large multispeciality urological unit. Materials and methods From the clinical records, the diagnosis and procedure codes were ascribed to 106 finished consultant episodes (FCEs) in urology, by two urological trainees. The codes were compared with those ascribed by professional hospital coders land of which the trainees were unaware) from information written on the audit form by junior medical staff. Where there were discrepancies in codes an error was recorded and the stage in the coding process in which it occurred was determined. Results Forty-eight coding errors were found in 38 of the 106 (36%) FCEs;34 (71%) were caused by inaccurate coding and 14 (29%) were the result of the incorrect completion of audit forms, Conclusion The clinical codes generated from the authors' department do not accurately reflect the clinical practice. If coding errors of this magnitude are typical of urology units in general, the concept of hospital performance tables (which will be generated using routine clinical data) is untenable unless data recording is given higher priority.
Objective To summarize the Urological procedures recorded in Hospital Episodes Statistics (HES. 1998-99) and to examine the accuracy of coding and HES. Materials and methods Data on the 10 commonest urological procedu...
详细信息
Objective To summarize the Urological procedures recorded in Hospital Episodes Statistics (HES. 1998-99) and to examine the accuracy of coding and HES. Materials and methods Data on the 10 commonest urological procedures from the Department or Health website (***,.uk/hes) were extracted, summarized and presented. Results Urethral catheterization, endoscopic procedures on the bladder, prostate, urethra and ureter (excluding ureteric stone extraction), minor open procedures on the foreskin and the vas, bladder instillation, extracorporeal shockwave lithotripsy and prostatic biopsy are the 10 commonest procedures, according to finished consultant episodes. There is published evidence that the data from coding and HES are not completely accurate. Conclusion This study highlights shortfalls in the HES data (1998-99) which may directly affect the funding of urological services.
暂无评论