以大數據架構為基礎建置程式效能分析系統

程式執行效能的品質，是程式使用者滿度的重要指標之一，當發生程式執行效能問題時，程式開發人員為了找出影響程式效能的關鍵因素，需要藉由程式記錄檔所記載的資訊，對程式進行除錯及問題分析，讓程式恢復正常運作，以滿足使用者對程式效能的需求。對於龐大的MES（Manufacturing Execution System, 製造執行系統）系統架構而言，程式記錄檔不僅儲存位置分散、檔案數量及容量多，且程式記錄檔通常為非結構化格式，增加了程式開發人員的對記錄檔的收集、格式解析及數據分析的難度。大數據對於記錄檔的管理分析提供了許多技術及解決方案的架構，本研究將探討適合實際半導體晶圓廠MES的大數據技術架構，並在不影響原MES系統運作下，以大數據架構建置符合MES的程式效能分析系統，並找出已存在的程式效能問題。經本研究的程式效能分析系統的導入，程式發開人員將MES中的「查詢可生產晶圓」功能，由原本平均執行時間9秒，降低至2.3秒，有效提升MES系統的使用者滿意度。

關鍵字

大數據；程式；記錄檔；效能分析

並列摘要

Quality of software program performance is a key indicator of user satisfaction. When a performance problem occurs, software developers must restore normal operation by identifying the key factors behind the problem and debugging using log files. In the complex architecture of manufacturing execution systems (MESs), log files are stored in decentralized locations, are numerous in number, and feature substantial amount of data content. Moreover, log files are typically unstructured documents, which adds to the difficulty of data collection, parsing, and data analysis. Big data offer numerous technical frameworks and solutions for log file management. This study examined some of the frameworks that are suitable for practical semiconductor fab MES. We built an MES-compliant software program performance analysis system under the big data framework without affecting existing MES operations and identified existing performance issues. Using the software program performance analysis system proposed in this study, the average time for executing the “wafer to be produced ” query in the MES can be shortened to 2.3 seconds from 9 seconds, which significantly improves the MES with respect to user satisfaction.

並列關鍵字

Big Data ； Program ； Log File ； Performance Analysis

參考文獻

．Chen, M., Mao, S. and Liu, Y., (2014), Big data: a survey, mobile netw appl, Pages: 171-209, DOI:10.1007/s11036-013-0489-0.

．Fucheng, P., Haibo, S. and Bin, D., (2015), Manufacturing execution system present situation and development trend analysis, Information and Automation, 2015 IEEE International Conference on, DOI:10.1109/ICInfA.2015.7279345.

．Galbraith, J. R., (2014), Organization design challenges resulting from big data, 2014 by Organizational Design Community, DOI:10.7146/jod.8856.

．Laney Doug, (2001), “3D data management: controlling data volume, variety and velocity”, 2001 MetaGroup research publication.

．Sagiroglu, E., Sinang, D., (2013), Big data: a review, Collaboration Technologies and Systems (CTS), 2013 International Conference on, DOI:10.1109/CTS.2013.6567202.

國際替代計量

以大數據架構為基礎建置程式效能分析系統

全文下載

主題瀏覽