`
文章列表
进行数据解析的理由不计其数,相关的工具和技巧也同样如此。但是,当您需要用这些数据做一些新的事情时,即使有“合适的”工具可能也是不够的。这一担心对于异类数据源的集成同样存在。用来做这项工作的合适工具迟早应该是一种编程语言。 Oracle 提供了一些非常强大的实用程序来加载、处理和卸载数据。SQL*Loader、Data Pump、外部表、Oracle Text、正则表达式都能提供这些功能。然而人们常常会需要在数据库外做一些事情(或者,说得琐碎些,可能您还没有获得必要的数据库权限)。 利用 Python 可以进行高水平的、有效的数据解析。而利用互联网上免费提供的大量标准库和众多模块可以处理数据 ...
从 Python 2.4 版开始,cx_Oracle 自身可以处理 DATE 和 TIMESTAMP 数据类型,将这些列的值映射到 Python 的 datetime 模块的 datetime 对象中。因为 datetime 对象支持原位的运算操作,这可以带来某些优势。内置的时区支持和若干专用模块使 Pyt ...
在 Python 做事方式的核心原则中,有一个规定是要求具有到 API 的高级接口。数据库 API(在此例中为 Oracle API)就是一个例子。使用 Computronix 的 cx_Oracle Python 模块,您可以在维持与 Python 数据库 API 规范 v2.0 的兼容性的同时,控制 Oracle 的查询模型。 对于所有遵循该规范的客户端库而言,使用 DB API 2.0 查询数据库的模型都是一致的。在此基础上,cx_Oracle 的主要开发人员 Anthony Tuininga 添加了一组丰富的属性和方法,以向开发人员揭示 Oracle 独有的特性。仅用标准的方法而忘掉 ...
故障报错 Thread 1 cannot allocate new log, sequence 2594 Checkpoint not complete 故障现象 redo log频繁切换,数据库DML性能下降。额外的日志归档影响IO性能。 故障原因及解决方案 该故障大部分情况下是由于redo log切换过于频繁导致没有及时进行日志归档。此外,日志切换前没有完成检查点操作也会导致同样报错,可以通过调整fast_start_mttr_target参数优化检查点操作。 通过如下脚本可以检测出redo log的切换频率  SELECT
Oracle官方并不推荐在数据库物理结构发生改变的情况下进行Flashback,但有些特殊情况。例如项目上线时出现异常需要快速的回滚到表空间被删除之前的操作。那么flashback tablespace想比于进行全库的restore和recover能更快速的满足要求,减少宕机时间。 具体演示 SQL> create tablespace tt datafile '/home/oracle/tt.dbf' size 10m;Tablespace created.SQL>  conn hr/hr;Connected.SQL> create table t1 tablespac ...
Question:  I need to monitor a data pump export job and I want to monitor the progress of the expdp command.  What are the ways to monitor a Oracle export job?   Answer:  The data pump logfile is the best way to monitor the progress on a data pump export, as data pump writes it's progress on a reg ...
#精确计算表空间大小,消耗系统资源,慎用SELECT F.TABLESPACE_NAME,        A.ALL_TOTAL "总空间" ,        A.ALL_USED "总使用空间" ,        A.ALL_TOTAL - A.ALL_USED "总剩余空间" ,        (A.ALL_TOTAL - A.ALL_USED) / ...
With the advent of the /*+ APPEND_VALUES*/ hint in 11gR2, I suspect we will see the feature misused and a bunch of questions on why it "doesn't work". The documentation states that "direct-path INSERT can be considerably faster than conventional INSERT.". What it should state is ...
The Four Step Program to Using utPLSQL Step 1. Install utPLSQL. Step 2. Choose a program to test and identify the test cases. Step 3. Build a test package. Step 4. Run your test. A note on which schemas to use Where to go from here Step 1. Install (and Upgrade) utPLSQL. Note: if you hav ...
Rich Set of DimensionsDimensions provide context for facts. Without context, facts are impossible to interpret. For example, I might tell you, “Order dollars are $40,000.” This statement is of no use to you. It presents a fact (order dollars) without any explanation. Is this orders for one product ...
Type 3 ChangesIn most dimensional schemas, the bulk of changes to source data generate type 1 and type 2 changes. Occasionally, neither technique satisfies. A third type of change response is called into play when there is a need to analyze all facts, those recorded before and after the change, with ...
The Hybrid ResponseWhen requirements call for changed data to result in both type 1 and type 2 behavior, the solution is disarmingly simple: provide for both. This cannot be achieved with a single attribute, but it can be achieved by providing a pair of attributes. Each attribute has the same sourc ...
大多数的锁都是行级锁,它们防止两个会话修改相同的行。事务的ACID(原子性、一致性、隔离性和持久性)属性,特别是一致性,它要求Oracle限制对表数据的并发修改。锁(lock)机制用于管理对共享资源的并发访问,Oracle也会 ...
Time-Stamped DimensionsIf there is any uncertainty about requirements for historic data, the most common response to changes in source data is the type 2 slowly changing dimension.It is the safe choice because it preserves the association of historic dimension values with facts that have been record ...
Slowly Changing Dimensions The "Slowly Changing Dimension" problem is a common one particular to data warehousing. In a nutshell, this applies to cases where the attribute for a record varies over time. We give an example below: Christina is a customer with ABC Inc. She first lived in ...
Global site tag (gtag.js) - Google Analytics