200905497 j 九、發明說明： 【發明所屬之技術領域】 本發明係關於一種資料倉儲品質檢核系統'，特別是指一種即時戍定期 谓測資料庫内資料内容異常情形，記錄並能自動回報給特定使用者之系统。 【先前技術】200905497 j IX. INSTRUCTIONS: [Technical field of invention] The present invention relates to a data storage quality inspection system, in particular to an abnormal situation of data content in a real-time periodic survey database, which is recorded and automatically returned to System for a specific user. [Prior Art]
隨著資料倉儲資料量逐日快速成長，資料進入資料倉儲品質之良窥關 係到整個資料倉儲成敗。以往資料進Λ資料倉儲前，經由資料擷取、轉換 及載入(Extract Transformation and Loading，ETL)過程可以篩選出不符人才夂 式定義的資料，但此僅能去除資料型態不符合部分，對於資料格式相同但 資料内容因使用者冑入錯豸、傳輸過程失真及資料拮脉時程式撰寫錯誤等 因素造成資料異常則無法得知，該類異常資料進人資料倉儲後，日久易造 成資料分析可雜降低’進而導致可能會造錢續資料分析結果異常、決 策la决導致整個資料倉儲系統的失敗’對於動軋以千萬計之投資成本而 δ，改進資料倉儲資料品質實為當務之急。 由此可見，上述制方式仍有諸多缺失，實非—良善之設計，而封寺. 加以改良。 本案發明人鑑於上述習用方式所触的各項缺點H思加以改良創 新’並經多年苦心職潛心研，終於成功研發完成本件龍倉儲品質 檢核系統。 貝 【發明目的】 本發明之目的即在於提供料倉™«統，齡資料庫端 提供—系列資料品質規則及條件.，用以規範所限定之資料内容，用以即時 200905497 或定時伯測資料倉儲内資料品質 、立依顧^较檢歸件，自動及定 期產生檢核報表給特定使用者 便用者了以根據檢核報表判斷 料進行維護，用以維護資料倉儲 、 芦曼貝针讀内㈣的正確性，使f料倉儲資料品質可 以維持於一定水準，避免後續使 、期如T分析的人員因品質不良之資 料作出不正確之分析結果。 ' 本發明之-人-目的係在於提供—種資料倉儲品質檢核系統，該系统可 以隨時新增及糊瀬罐轉_，_f f糊的速度。 【發明内容】 11達成上述發明目的之資料倉儲品質檢核系統，侧用資料庫系統内 建之edPiOeedure即日该定期侧資料相資湘容異錄形，依預先 定義之資料品質條件進行資卿檢；本發明包含以下模組： 介面異動/刀析柄組，主要依預先設定規則及預設關聯資料找出異常 介面及後齡糾料之資觸取、賴及心(etl)作業； - ETL作業錯誤分析處理模組，於f料進人資料倉儲過程所產生異常 資料之;7析力I模組，產生異常分析報表作為資料擷取、轉換及載入讲几) 作業改善使用； . -貝料定航對她，依據贱設定制分別輯來狀資料倉儲資 料產生異常報表，作為資料一致性維護使用； -規則貧料庫，係㈣儲存使用者定義之資料品質檢核規則； -條件貢料庫’係用以儲存使用者定義之該規麟需符合的條件； 一兀貢料(Metadata) ’係用以記錄載入資料倉儲檔案格式、使用到的資 200905497 料娜、轉換及載人(ETL)轉換程錢載人資料倉儲資料表； -報表服務，侧„料賴供謂存程序⑼⑽他⑽㈣掘取 資料庫内相關資料產生所需之異常報告；該報表服務模組係作為該介面異 動分析模組、輸輸取、龍及以购作細_處賴組，以及 該資料定期比對模組之前端顯示介面； 一異常分㈣務餘，侧帛龍賴供之辦賴㈣ -p預先疋義㈣異倾型進人賴及評量資料，透過資觸取、轉換及 載入(ETL)過程進人資__料品#水準，將異常結果_來供使用 ^考使用〜亥異吊分析服務模組需與該資料擷取、轉換及載入(飢)作業 錯誤分析處理模組連結，以取得異常資料來進行異常分析；以及 ”吊D S 貝料庫’係作為異常分析結果儲存結果的目的資料庫，該 異常分析職漁所得分赌果係織在《常告«料料，'而該介面 、刀析模、、且及資料定期比對模組比對後結果也會儲放到該異常告警資料 庫中。 e 使用者可事蚊義資料品f檢核規則於該規則資料庫，另定義規則所 需符合的條件於該條件資料庫，另需要載入資料倉儲檔案格式、使用到的 貧料操取、轉換及載入(ETL)轉換程式及載入資料倉儲資料表皆需記錄於該 兀資料(Metadata)内；使用者在資料進入資料倉儲前可先進行介面比對，根 f該元__adata)所齡f料，透麟介面魏分析漁，比對來源及 資料倉儲間攔位、資料長度及龍鶴預先進行異㈣f ;當資料於資料 擷取、轉換及載入(ETL)過程會因不符轉換規則、型態不符等因素外無法進 200905497 入資料倉麟’可透職資觸取、雜及載人(ETL)作業錯誤分析處理模 組，根據賊職料庫所預先定義制，並配合歸件賴庫所定義可容 許錯誤範圍，進行分減出錯誤原因並_給㈣來源或修JL資料掏取、 轉換及載人(ETL)程式；tf料進料倉織，亦可透過該資料補比對 模組進仃貞料品質監測及分析，同樣也是根據減則資料庫所預先定義之 規則’以及配合該條件賴庫所定義可料錯誤顧，找出與企業規則定 義不符之《料，並分析其進人㈣倉歡原因，並透過該報表服務模組利 用Stored Procedure產生異常報告，且使用該異常分析服務模組配合As the amount of data stored in the warehouse grows rapidly day by day, the data enters the quality of the data storage and the quality of the data storage success or failure. Before the data is imported into the data warehouse, the data can be filtered out by the Extract Transformation and Loading (ETL) process. However, this can only remove the non-conformity of the data type. The data format is the same, but the data content is not known due to factors such as user error, distortion of transmission process, and incorrect writing of data when the data is written. This type of abnormal data is easily stored after being stored in the data. The analysis can be mixed and reduced, which leads to the possibility of making money and the abnormal results of data analysis, and the failure of the decision to lead to the failure of the entire data storage system. For the investment cost of tens of millions of tens of millions of operations, improving the quality of data storage is a top priority. It can be seen that there are still many shortcomings in the above-mentioned system, which is not the design of goodness, but the temple is to be improved. In view of the shortcomings of the above-mentioned conventional methods, the inventor of the present invention improved and innovated, and after years of painstaking research, finally successfully developed and completed the quality inspection system for the dragon storage quality. [Object of the Invention] The object of the present invention is to provide a material warehouse TMTM system, a database of quality standards and conditions for the specification of the data, for the purpose of standardizing the information content, for instant 200905497 or timed data. The quality of the data in the warehouse, the inspection according to the inspection, the automatic and regular inspection reports are issued to the specific users, and the maintenance is carried out according to the inspection report, to maintain the data storage, the Aumanbei needle reading The correctness of the internal (four), so that the quality of f material storage data can be maintained at a certain level, to avoid the subsequent analysis, such as T analysis personnel due to poor quality of the data to make an incorrect analysis results. The invention - the purpose of the invention is to provide a data storage quality inspection system, which can be added at any time and can be used to change the speed of the _, _f f paste. [Summary of the Invention] 11 The data storage quality inspection system that achieves the above-mentioned object of the invention, the edPiOeedure built into the side database system, and the regular side data of the current phase of the Xiangxiang different records, according to the pre-defined data quality conditions for the capital inspection The invention comprises the following modules: interface change/knife handle set, mainly according to preset rules and preset related data to find out the abnormal interface and the post-age correction material access, and the heart (etl) operation; - ETL The operation error analysis and processing module generates abnormal data generated during the data storage process of the material; 7 analysis of the force module, generates an abnormal analysis report as data acquisition, conversion and loading.) According to the 贱 setting system, the data is stored in the warehousing data to generate an abnormal report, which is used as data consistency maintenance; - the rule of the poor library, the (4) storage user-defined data quality inspection rules; - condition tribute The library ' is used to store the user-defined conditions for the ruler; the Metadata is used to record the data file format and the used resources 20090549 7 纳娜, conversion and manned (ETL) conversion process money manned data warehousing data sheet; - report service, side _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ The report service module is used as the interface transaction analysis module, the input and output, the dragon and the purchase as the fine-going group, and the data is periodically compared to the front end of the module display interface; an abnormality (four) surplus, side帛 赖 赖 赖 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( Results _ for use ^ test use ~ Haiyi crane analysis service module needs to be linked with the data capture, conversion and loading (hungry) error analysis processing module to obtain abnormal data for abnormal analysis; and "hanging The DS Shell Library is the destination database for the results of the abnormal analysis results. The anomaly analysis of the fishermen's scores is based on the "consultation material," and the interface, the knife, and the data are regularly After the comparison module is compared, the result will be stored in the abnormal alarm data. In the library. e The user can use the rules of the mosquito data to check the rules in the rule database, and define the conditions that the rules need to meet in the condition database, and also need to load the data storage file format, use the poor materials, Both the conversion and loading (ETL) conversion program and the loaded data storage data sheet need to be recorded in the metadata (metadata); the user can perform the interface comparison before the data enters the data storage, and the root f is the element __adata) The age of f material, through the Lin Wei Wei analysis of fishing, comparison of source and data storage block, data length and Longhe pre-existing (four) f; when the data in the data acquisition, conversion and loading (ETL) process will not match In addition to the conversion rules, the type of non-conformity and other factors can not enter the 200905497 into the data warehouse Lin 'capable of the employment, miscellaneous and manned (ETL) operation error analysis and processing module, according to the thief job library pre-defined system, and with The scope of the error allowed by the dependent library is to reduce the cause of the error and to (4) source or repair JL data acquisition, conversion and manned (ETL) procedures; tf material feed warehouse, through this information Replenishment monitoring of the quality of the module The analysis is also based on the pre-defined rules of the deduction database and the misunderstandings defined by the requirements of the library, to find out the material that does not conform to the definition of the enterprise rules, and analyze the reasons for entering the people (four) Cang Huan, and Use the Stored Procedure to generate an exception report through the report service module, and use the exception analysis service module to cooperate
Procedure將異常結果篩選出來供使用者參考使用，紐由使用者2決定是 否需要修正。 【實施方式】 本發明係為提昇資料擷取、轉換及載入(ETL)作業資料品質，以確保資 料倉儲内㈣品獅持在可接受水準。在麟倉射資侧取、轉換及載 入(ETL)作業中確保高品質之㈣進人龍倉儲為—必要的ϋ參閱圖 一，本發明所提供之資料倉儲品質檢核系統i，主要包括有： 一介面異動分析· U :主要依預先設定規則及預設關师料找出異 常介面及後續會受到影響之資料擷取、轉換及載人(飢)作# ; -資料齡、雜及.(ETL)健錯誤分域雜組12 :於資料進入 資料倉儲過程所產生異常f料之分析功賴組，產生異常分析報表作為資 料擷取、轉換及载入(ETL)作業改善使用； 1料定期比對模組13 :依據預先設定規則分別比對來源及資料倉儲. 200905497 資料產生異书報表，作為資料一致性維護使用； 規則貝料庫14 ·係用以儲存使用者定義之資料品質檢核規則； 條件貝料庫I5 .係用以健存使用者定義之該規則所需符合的條件； 兀資料(Metadata) Ιό ·係用以記錄載入資料倉儲檔案格式、使用到的 資料操取、雜及載人(ETL)轉触式及•資料纽資料表； 報表服務馳17，係细倾庫提供讀絲序(s_d p_dure) 擷取資料庫内相關資料產生所需之異常報告；該報表服務模組係作為該介 面異動分㈣組11、«觸取、轉換及·_獅錯誤分析處理模組 12，以及«料定誠對模組13之前端顯示介面； 異吊分析服務馳18，係—f料庫提供之儲存程序(Stored Procedure)配合預先定義資料異常模型進入篩選及評量資料，透過資料擷 轉換及載入(ETL)過私進入資料倉儲的資料品質水準，將異常結果筛選 -吏用者參考使用，韻常分析服務模組需與該資棚取、轉換及載 讀錯齡析處賴組叫結，輝得異饱槪減常分析； 以及 > #、* °。貝料庫19 ’係作為異常分析結果儲存結果的目的資料庫， «吊刀析服務模組18所得分析結果_放在該異常告”料庫β中， 而該介面異動分析模組11及資料定期比對模組η比對後結果也會儲放到 該異常告警資料庫19中。 :所7^ ’使用者2可事先定義資料品質檢核規則於該規則資料庫 另疋義規則所需付合的條件於該條件資料庫Μ，另需要載入資料倉儲 200905497 檔案格式、使用到的資料擷取、轉換及載入(ETL)轉換程式及載入資料倉儲 資料表皆需記錄於該元資料(Metadata) 16内，資料倉儲品質檢核系統丨會透 過該報表服務模組17利用Stored Procedure產生異常報告，並使用該異常, 分析服務模組1S配合Stored Procedure將異常結果篩選出來供使用者2參 考使用。 Μ者2在資料進人資料倉儲5前可先進行介面比對，根據該元資料 (Metadata) 16所儲存資料，透過該介面異動分析模組η，比對來源及資料 〇倉儲間攔位、資料長度及資料型態預先進行異常示警；當資料於資料掏取、 轉換及載人(ETL)過程赖不符轉換删、型態不符軸素外無法進入資料 倉儲時’可透職資觸取、轉換域人(ETL)作業錯誤分減理模組Η， 根據該規卿_ Η所預先定義綱，並配合該條件資料庫b所定義可 合許錯决範圍’進行分析找出錯誤原因並回鑌給資料來源3或修正資料擷 取、轉換及載人(ETL)程式4 ·，t㈣進人資料倉儲5後，亦可透過該資料 (定期比對模組η進行資料品質監測及分析，同樣也是根據該規則資料庫14 斤預先定義之規則’以及配合該條件資料庫U所定義可容許錯誤範圍，找 出與企業規則定義不符之資料，並分析其進人資料倉儲之原因，再由使用 者2決定是否需要修正。 、_ π參關―，本發明之介面異動分析模組的運作流細如圖二所示， 當㈣來源端與資料倉健間所定義的資料格式有所異動時 ，會啟動該介面， 異動刀析模組’由制者輸入舖後之介面格錢，該難會讀取前次儲 子之面格式’透過預先定義於規則資料庫内之比對規則，將使用者輸入 200905497 異動後之介祕式與前次儲存之介面格式進行比對，若輯結果顯示前後 )|面格式不致時，再進-步對元資料(Metad㈣内所儲存之關聯，進一步 進浦擊f彡響分析，找出时喊動___相_卿取、轉換及 載入(ETL)程式，並產生介面異動報告，以及記錄介面規格比對結果，以提 供使用者參考；若比對結果黯錢介面格式—輯，則賴組直接記錄 介面規格比對結果，以提供使用者參考。The Procedure filters the abnormal results for the user's reference, and the user 2 determines whether the correction is needed. [Embodiment] The present invention is to improve the quality of data acquisition, conversion and loading (ETL) operation data to ensure that the lion in the data storage is at an acceptable level. In the Lincang side-by-side acquisition, conversion and loading (ETL) operation to ensure high quality (4) into the dragon storage - necessary ϋ see Figure 1, the data storage quality inspection system i provided by the present invention, mainly includes Yes: One interface analysis · U: Mainly based on pre-set rules and presets to find out the abnormal interface and subsequent impacts on data acquisition, conversion and manned (hunger) #; - data age, miscellaneous (ETL) Health Error Domain Miscellaneous Group 12: The analysis of the abnormality of the material generated in the data storage process, the abnormal analysis report is generated as the data acquisition, conversion and loading (ETL) operation to improve the use; Periodic comparison module 13: Compare the source and data storage according to the pre-set rules. 200905497 Data generated different book report, used as data consistency maintenance; Rule library 14 is used to store user-defined data quality Checking rules; conditional shell library I5. is used to store the user-defined conditions required by the rule; 兀Metadata Ιό ·Used to record the data storage file format, used data take, And the manned (ETL) touch-type and • information information table; the report service Chi 17, the fine-grained library provides the reading order (s_d p_dure) to retrieve the relevant information in the database to generate the required abnormal report; the report service The module system serves as the interface change group (4) group 11, the «touch, conversion and _ lion error analysis processing module 12, and the material display interface of the front end of the module 13; the different crane analysis service Chi 18, is - The Stored Procedure provided by the f-series cooperates with the pre-defined data anomaly model to enter the screening and assessment data, and filters the abnormal results through the data quality conversion level of the data storage and loading (ETL). For reference, the rhythm analysis service module needs to be associated with the arbitrage, conversion, and misplaced analysis of the stagnation, and the analysis of the suffocation; and >#, * °. The shell library 19' is used as the destination database for the results of the abnormal analysis results storage, «the analysis result of the hanging knife analysis service module 18 is placed in the abnormality report library β, and the interface transaction analysis module 11 and the data The results of the regular comparison module η are also stored in the abnormal alarm database 19. The user 2 can define the data quality check rule in advance in the rule database. The conditions for the reconciliation are in the condition database, and the data storage warehousing 200905497 file format, the data retrieval, conversion and loading (ETL) conversion program and the loading data warehousing data sheet are all recorded in the quotation. In the data (Metadata) 16, the data storage quality checking system will use the stored procedure to generate an abnormal report through the report service module 17, and use the abnormality, and the analysis service module 1S cooperates with the stored procedure to filter the abnormal result for the user. 2 Reference use. The second person can compare the data before entering the data storage 5, according to the data stored in the metadata (Metadata) 16 through the interface transaction analysis module η, the comparison Source and data 〇 warehousing block, data length and data type are pre-warned; when the data is extracted, converted and manned (ETL) process does not match conversion, type does not conform to axis, can not enter data warehousing At the time of 'receiving the job, the transfer domain (ETL) operation error sub-reduction module, according to the pre-defined outline of the _ _ , 并, and with the definition of the condition database b can be agreed to the scope of the wrong decision 'Analysis to identify the cause of the error and return to the source 3 or the revised data acquisition, conversion and manned (ETL) program 4 ·, t (4) into the data storage 5, can also pass the information (regular comparison module η data quality monitoring and analysis, also according to the rules of the library 14 kg pre-defined rules 'and the tolerance range defined by the condition database U, to find out the data that does not conform to the definition of the enterprise rules, and analyze its progress The reason for the warehousing of the person is determined by the user 2 as to whether or not the correction is required. _ π 参 ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ Place When the data format of the righteousness changes, the interface will be activated. The template of the fixed-action tool is 'input by the maker, and the format of the previous store's face is read.' The comparison rule in the library compares the secret type of the user input 200905497 after the transaction with the previous storage interface format. If the result of the series is displayed before and after the | face format is not met, then the step-by-step metadata (Metad(4) The correlation stored in the library is further analyzed by the sniper, and the ___phase_clearing, converting and loading (ETL) program is generated, and the interface transaction report is generated, and the interface interface comparison result is recorded. To provide user reference; if the comparison results in the interface format, the Lai group directly records the interface specification comparison results to provide user reference.
V 請參閱圖三’本發明之資料掏取、轉換及載入(etl)作業錯誤分析處理 模組的運作流程圖如圖三所示，當資料自f料來源端要載人資料倉儲前， 必需要先驗«_容及格式，並絲不符合介面格式資料，再經過企業 邏輯轉換篩選出符合需求之資料内容，並加上資料倉儲必需要之資料，最 後再載入倾倉儲，上述三個步驟皆可能產生異常㈣，為提高資料進入 資料倉儲之品質’就需對此異料料進行分析；·，本發明之系統會啟 動該資料娜、概及載人(ETL)作業錯齡減賴組，個預先定義於 規則資料庫巾之檢驗酬’並配合條件龍庫喊義之條件範圍資料，對 無法符合資料擷取、轉換及載入(ETL)轉換規則之資料進行檢驗，找出異常 原因進行修正；首先，將擷取資料及來源資料進行驗證，比對擷取資料來 源是否正常，若正常則進行資料轉換，若異常則進行異常資料處理、異常 分析，以及提供異常報表；進行資料轉換後之資料則會繼續被分析是否符 合轉換規格，右符合則進行資料載入，若異常則進行異常資料處理、異常 分析，以及提供異常報表；進行資料栽入之資料則會繼續被分析載入是否 正常，若正常則結束本模組分析流程，若異常則進行異常資料處理、異常 200905497 分析，以及提供異常報表β 次請參_四，本發明之資料定期比對模_運作流程圖如_所示， 貝枓進入貝枓倉儲後，累積—段時間後可能產出資料重複、缺漏及異常等 問題，此時就需與來«進行_同步轉持動卜雜；因此，本發明 之系統會啟動該資料定期比對模組，利用預先定義規則資料庫内比對規 則，並配合齡諸庫岐義之條件顧，對麵端及錯内資料進 灯輯’首先’擷取比對來源資料，並讀取預先定義比對規則，進行比對 原有貝料觸貞料倉儲㈣錄是碑合該崎删，若符合該比對規 丨則繼’賣進订了筆規則之比對此時模組會判斷是否有其他規則要比 U本触分析雜，若有則本模組會重新讀取預蚊義比對 丨找出需要比對的規則繼續比對；至於比對出不符合該比對規則的資 料本模組則產生出異常報表供使用者進行資料校正。 【特點及功效】 本毛月所提t、之資料倉儲品質檢核系統，與其他習用技術相互比較 時，更具備下列優點.· 1. 本發明可將資料品f具舰，提供高f4#料品質。 2. 本發明可將資料品f工作集中控管，提昇f理效率。 3·本發明可由使用者自訂品質項目儲存於資料倉儲資料庫内，隨時可 新增及修改酬提高品之可雜及靈活度。 上列詳細說明乃針對本發明之一可行實施例進行具體說明，惟該實施 例並非用㈣制本發明之專植圍，凡未麟本發難藝餅所為之等效 12 200905497 實施或變更’均應包含於本案之專纖圍中。 述本案不僅於技術思想上確屬創新，並具備習用之傳統方法 <述夕項功效’已充分符合新穎性及進步性之法定發明專利要 件’友依法提出中請，_請t域准本件發日轉利巾請案，以勵發明， 至感德便。 【圖式簡單說明】 ' —請參閱有關本㈣之詳細綱及其_，將可進—步瞭解本發明之技 Ο術内容及其目的功效；有_圖為： 圖—為本發明資料倉儲品質檢核系統之架構圖； 圖二為該資料倉儲品f檢核纽之介面购分析模域糊； 圖三為該資料續品質檢⑽統之龍擷取、娜及載⑽叫 誤分析處理模組流程圖；以及 一曰 圖四為該資料倉儲品質檢核系統之資料定期比對模組流程圓 【主要元件符號說明】 11 12 15 ΐ料倉储品#檢核系統 介面異動分析模組 理模組 規則資料庫 條件資料庫 資料擷取、轉換及載入(ETL)作業錯誤分析處 資料定期比對模組 W 元資料(Metadata) Π報表服務模組 Μ異常分析服務模組 13 200905497 19異常告警資料庫 2使用者 3資料來源 4資料擷取、轉換及載入(ETL)程式 5資料倉儲V Please refer to Figure 3 'The data flow chart of the data acquisition, conversion and loading (etl) operation error analysis and processing module of the present invention is shown in Figure 3. Before the data is sourced from the source of the material, It is necessary to a priori «_ tolerance and format, and does not meet the interface format data, and then through the enterprise logic conversion to filter out the information content that meets the requirements, plus the information necessary for data storage, and finally load the dump storage, the above three Any step may cause anomalies (4). In order to improve the quality of data entering the data storage, it is necessary to analyze the different materials. The system of the present invention will initiate the data, and the ETL operation is less than the age. Lai group, a pre-defined test fee for the rule database, and with the conditions of the conditions of the Longku shouting, the data that cannot meet the data acquisition, conversion and loading (ETL) conversion rules are tested to find the abnormality. The reason is corrected; firstly, the data and source data are verified, and the source of the data is compared. If it is normal, the data is converted. If abnormal, the abnormal data is processed. Often analyze and provide abnormal reports; the data after data conversion will continue to be analyzed to meet the conversion specifications, the right is to load the data, if abnormal, the abnormal data processing, abnormal analysis, and provide abnormal reports; The data that is planted will continue to be analyzed and loaded normally. If it is normal, the analysis process of this module will be terminated. If abnormal, the abnormal data processing, abnormality 200905497 analysis, and the abnormal report β times will be provided. The data is regularly compared with the model _ operation flow chart as shown in _, after the 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 , , 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓 枓Therefore, the system of the present invention activates the periodic comparison module of the data, utilizes the matching rules in the pre-defined rule database, and cooperates with the conditions of the ages of the library, and enters the data on the opposite side and the wrong data. The series 'first' captures the source data and reads the pre-defined comparison rules to compare the original beetles and warehousing (four) records. If the code is in compliance with the rules, then the module will judge whether there are other rules than the U touch. If there is any, the module will re-read. The pre-mosquito ratio is used to find out the rules that need to be compared. The data module that does not meet the comparison rule produces an abnormal report for the user to correct the data. [Features and Efficacy] The data storage quality inspection system of the t-data of the month is better than the other conventional technologies. The invention has the following advantages: 1. The invention can provide the data item f ship with high f4# Material quality. 2. The invention can centrally control the work of the data item f, thereby improving the efficiency of the material. 3. The invention can be stored in the data storage database by user-defined quality items, and the complexity and flexibility of the rewards can be added and modified at any time. The above detailed description is specific to a possible embodiment of the present invention, but the embodiment does not use (4) to make the special planting of the present invention, and the equivalent of 12, 200905497 is implemented or changed. It should be included in the special fiber enclosure of this case. The case is not only innovative in terms of technical thinking, but also has the traditional method of learning. The effect of the syllabus has been fully in line with the novelty and progressiveness of the statutory invention patents. Sending a case to the towel, in order to invent the invention, to the sense of virtue. [Simple description of the schema] ' - Please refer to the detailed outline of this (4) and its _, you will be able to further understand the technical content of the invention and its intended effect; there is _ picture: Figure - data storage for the invention Figure 2 is the structure diagram of the quality inspection system; Figure 2 is the interface analysis of the data warehouse 检 纽 ; ; ; ; ; 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图 图Module flow chart; and Figure 4 is the data storage quality check system data periodic comparison module flow circle [main component symbol description] 11 12 15 仓 仓 品 # 检 检 系统 系统 系统 系统 系统 系统 系统 异Management Module Rules Database Condition Database Data Acquisition, Conversion and Loading (ETL) Job Error Analysis Department Data Periodic Comparison Module W Metadata Π Reporting Service Module Μ Anomaly Analysis Service Module 13 200905497 19 Abnormal Alarm Database 2 User 3 Data Source 4 Data Capture, Conversion and Loading (ETL) Program 5 Data Warehousing
1414