Database history - Revision history

62.233.164.1 at 08:52, 29 December 2005

2005-12-29T08:52:38Z

MaciejBlizinski: /* The database structure */

2005-09-30T07:31:13Z

‎The database structure

MaciejBlizinski: /* The "BIG" table */

2005-09-30T07:30:19Z

‎The "BIG" table

MaciejBlizinski: /* The database structure */

2005-09-30T07:29:00Z

‎The database structure

MaciejBlizinski: /* The main ideas */

2005-09-30T07:27:17Z

‎The main ideas

MaciejBlizinski: /* The database structure */

2005-09-30T07:26:04Z

‎The database structure

MaciejBlizinski at 13:22, 24 September 2005

2005-09-24T13:22:45Z

MaciejBlizinski at 13:21, 24 September 2005

2005-09-24T13:21:59Z

MaciejBlizinski at 09:38, 24 September 2005

2005-09-24T09:38:08Z

MaciejBlizinski: /* The main ideas */

2005-09-24T08:12:50Z

‎The main ideas

@@ Line 19: / Line 19: @@
 == The history ==
-Up to summer 2003 the central database runned MySQL and it's structure was the same as the structure of the echdd6's. The problems with the referential integrity pushed us to more advanced DBMS, and our choice was PostgreSQL]]. We also re-designed the structure, and made it more modern and flexible.
+Up to summer 2003 the central database has run MySQL and it's structure was the same as the structure of the echdd6's. The problems with the referential integrity pushed us to more advanced DBMS, and our choice was PostgreSQL]]. We also re-designed the structure, and made it more modern and flexible.
 The data, when being imported, needed to be converted from old to new structure. There were a special piece of software that did it. The web interface avoided this problem, as it operated directly on the proper database structure.

←Older revision		Revision as of 07:31, 30 September 2005
Line 10:		Line 10:

	== The database structure ==		== The database structure ==
		+
		+	[[Image:DatabaseStructure.png\|thumb\|480px\|right\|The old (up to year 2004) database structure diagram]]
		+
		+	This diagram shows the database structure. It uses standard database structure notation, except for that there are no dashed lines here.

	''See the article: [[database structure]].''		''See the article: [[database structure]].''

@@ Line 65: / Line 65: @@
 == The "BIG" table ==
-Once upon a time, the computers were too slow to perform joins of many tables on the fly in acceptable amount of time. The programmers created a special join of several tables from the ECDB database and called it the "BIG" table. It was used, and it's still being used as the source of data for GoldStandards reports. Before, it was created from MySQL tables, which don't
+''outdated information:''
-The creation of the BIG table is a complex query, which involves many joins, conditional expressions and subqueries, and it's intended to produce exactly the same output as the former BIG-creating query. Unfortunately, due to its complexity, the output is only 99% the same, which means that there are always small differences between reports generated directly from PostgreSQL and reports generated from BIG.
+Once upon a time, the computers were too slow to perform joins of many tables on the fly in acceptable amount of time. The programmers created a special join of several tables from the ECDB database and called it the "BIG" table. It was used, and it's still being used as the source of data for [[Gold standards report]]s. Before, it was created from MySQL tables, which don't exist now, and the "BIG" table is emulated in PostgreSQL, and then copied to MySQL, from where the (some of the) online reports are served.
 '''2004 update''': This table doesn't exist anymore. All reports are now driven by [[PostgreSQL]] alone.

@@ Line 11: / Line 11: @@
 == The database structure ==
-[[Image:DatabaseStructure.png|thumb|480px|right|The database structure diagram]]
+''See the article: [[database structure]].''
 == The history ==

@@ Line 23: / Line 23: @@
 == The main ideas ==
-Formerly the diagnoses, procedures, GPRF and NCA were stored in separate tables and separate interconnection tables. In new database structure they were generalized to be ''factors''.
+Formerly the diagnoses, procedures, [[GPRF]] and [[NCA]] were stored in separate tables and separate interconnection tables. In new database structure they were generalized to be ''factors''.
 The idea of factors is that the given factor (in the database) is, or is not, connected with given object. For instance, patient P1 was or was not given the D1 diagnosis. In other words, for a given set of objects (e.g. patients) and a set of factors (e.g. diagnoses) there is a defined relation over the cartesian product of these two sets.
-Diagnoses, procedures, complications, GPRF and NCA became the factor classes. There is also a new class, called "aristo", which is intended to contain Comprehensive (Aristotle) Score data. As the classes are represented by the rows of the <tt>factor_class</tt> table, new classes can be introduced in future, without any need to redesign the database's structure.
+Diagnoses, procedures, complications, [[GPRF]] and [[NCA]] became the factor classes. There is also a new class, called "aristo", which is intended to contain Comprehensive Score ([[Aristotle Score]]) data. As the classes are represented by the rows of the <tt>factor_class</tt> table, new classes can be introduced in future, without any need to redesign the database's structure.
 == The dictionary tables ==

@@ Line 33: / Line 33: @@
 == The dictionary tables ==
-...have a "<pre>dict_</pre>" prefix, so they can be easily distinguished from the main tables. This is the listing of the database's tables:
+...have a "<tt>dict_</tt>" prefix, so they can be easily distinguished from the main tables. This is the listing of the database's tables:
 <pre>
@@ Line 80: / Line 80: @@
 The database undewent one more redesign. The '''admission''' was introduced as a new table in the database. More on this subject in [[User's manual]] of the [[EACTS Database Software]].
-[[category:terminology]]
+[[category:terminology|History, database]]

←Older revision		Revision as of 13:21, 24 September 2005
Line 79:		Line 79:

	The database undewent one more redesign. The '''admission''' was introduced as a new table in the database. More on this subject in [[User's manual]] of the [[EACTS Database Software]].		The database undewent one more redesign. The '''admission''' was introduced as a new table in the database. More on this subject in [[User's manual]] of the [[EACTS Database Software]].
		+
		+	[[category:terminology]]

@@ Line 7: / Line 7: @@
 = The database change in 2003 =
-The central database switched to the [[PostgreSQL]] RDBMS, which is the most advanced open source database engine.
+The central database switched to the [[PostgreSQL]] RDBMS, which is the most advanced [[open source]] database engine.
 == The database structure ==
@@ Line 14: / Line 14: @@
 ''The database structure diagram.''
 This diagram shows the database structure. It uses standard database structure notation, except for that there are no dashed lines here.