Search Results

Search found 36946 results on 1478 pages for 'sample database'.

Page 40/1478 | < Previous Page | 36 37 38 39 40 41 42 43 44 45 46 47  | Next Page >

  • Is it better to use a relational database or document-based database for an app like Wufoo?

    - by mboyle
    I'm working on an application that's similar to Wufoo in that it allows our users to create their own databases and collect/present records with auto generated forms and views. Since every user is creating a different schema (one user might have a database of their baseball card collection, another might have a database of their recipes) our current approach is using MySQL to create separate databases for every user with its own tables. So in other words, the databases our MySQL server contains look like: main-web-app-db (our web app containing tables for users account info, billing, etc) user_1_db (baseball_cards_table) user_2_db (recipes_table) .... And so on. If a user wants to set up a new database to keep track of their DVD collection, we'd do a "create database ..." with "create table ...". If they enter some data in and then decide they want to change a column we'd do an "alter table ....". Now, the further along I get with building this out the more it seems like MySQL is poorly suited to handling this. 1) My first concern is that switching databases every request, first to our main app's database for authentication etc, and then to the user's personal database, is going to be inefficient. 2) The second concern I have is that there's going to be a limit to the number of databases a single MySQL server can host. Pretending for a moment this application had 500,000 user databases, is MySQL designed to operate this way? What if it were a million, or more? 3) Lastly, is this method going to be a nightmare to support and scale? I've never heard of MySQL being used in this way so I do worry about how this affects things like replication and other methods of scaling. To me, it seems like MySQL wasn't built to be used in this way but what do I know. I've been looking at document-based databases like MongoDB, CouchDB, and Redis as alternatives because it seems like a schema-less approach to this particular problem makes a lot of sense. Can anyone offer some advice on this?

    Read the article

  • Database choices

    - by flobadob
    I have a prickly design issue regarding the choice of database technologies to use for a group of new applications. The final suite of applications would have the following database requirements... Central databases (more than one database) using mysql (myst be mysql due to justhost.com). An application to be written which accesses the multiple mysql databases on the web host. This application will also write to local serverless database (sqlite/firebird/vistadb/whatever). Different flavors of this application will be created for windows (.NET), windows mobile, android if possible, iphone if possible. So, the design task is to minimise the quantity of code to achieve this. This is going to be tricky since the languages used are already c# / java (android) and objc (iphone). Not too worried about that, but can the work required to implement the various database access layers be minimised? The serverless database will hold similar data to the mysql server, so some kind of inheritance in the DAL would be useful. Looking at hibernate/nhibernate and there is linq to whatever. So many choices!

    Read the article

  • Fragmented Log files could be slowing down your database

    - by Fatherjack
    Something that is sometimes forgotten by a lot of DBAs is the fact that database log files get fragmented in the same way that you get fragmentation in a data file. The cause is very different but the effect is the same – too much effort reading and writing data. Data files get fragmented as data is changed through normal system activity, INSERTs, UPDATEs and DELETEs cause fragmentation and most experienced DBAs are monitoring their indexes for fragmentation and dealing with it accordingly. However, you don’t hear about so many working on their log files. How can a log file get fragmented? I’m glad you asked. When you create a database there are at least two files created on the disk storage; an mdf for the data and an ldf for the log file (you can also have ndf files for extra data storage but that’s off topic for now). It is wholly possible to have more than one log file but in most cases there is little point in creating more than one as the log file is written to in a ‘wrap-around’ method (more on that later). When a log file is created at the time that a database is created the file is actually sub divided into a number of virtual log files (VLFs). The number and size of these VLFs depends on the size chosen for the log file. VLFs are also created in the space added to a log file when a log file growth event takes place. Do you have your log files set to auto grow? Then you have potentially been introducing many VLFs into your log file. Let’s get to see how many VLFs we have in a brand new database. USE master GO CREATE DATABASE VLF_Test ON ( NAME = VLF_Test, FILENAME = 'C:\Program Files\Microsoft SQL Server\MSSQL10.ROCK_2008\MSSQL\DATA\VLF_Test.mdf', SIZE = 100, MAXSIZE = 500, FILEGROWTH = 50 ) LOG ON ( NAME = VLF_Test_Log, FILENAME = 'C:\Program Files\Microsoft SQL Server\MSSQL10.ROCK_2008\MSSQL\DATA\VLF_Test_log.ldf', SIZE = 5MB, MAXSIZE = 250MB, FILEGROWTH = 5MB ); go USE VLF_Test go DBCC LOGINFO; The results of this are firstly a new database is created with specified files sizes and the the DBCC LOGINFO results are returned to the script editor. The DBCC LOGINFO results have plenty of interesting information in them but lets first note there are 4 rows of information, this relates to the fact that 4 VLFs have been created in the log file. The values in the FileSize column are the sizes of each VLF in bytes, you will see that the last one to be created is slightly larger than the others. So, a 5MB log file has 4 VLFs of roughly 1.25 MB. Lets alter the CREATE DATABASE script to create a log file that’s a bit bigger and see what happens. Alter the code above so that the log file details are replaced by LOG ON ( NAME = VLF_Test_Log, FILENAME = 'C:\Program Files\Microsoft SQL Server\MSSQL10.ROCK_2008\MSSQL\DATA\VLF_Test_log.ldf', SIZE = 1GB, MAXSIZE = 25GB, FILEGROWTH = 1GB ); With a bigger log file specified we get more VLFs What if we make it bigger again? LOG ON ( NAME = VLF_Test_Log, FILENAME = 'C:\Program Files\Microsoft SQL Server\MSSQL10.ROCK_2008\MSSQL\DATA\VLF_Test_log.ldf', SIZE = 5GB, MAXSIZE = 250GB, FILEGROWTH = 5GB ); This time we see more VLFs are created within our log file. We now have our 5GB log file comprised of 16 files of 320MB each. In fact these sizes fall into all the ranges that control the VLF creation criteria – what a coincidence! The rules that are followed when a log file is created or has it’s size increased are pretty basic. If the file growth is lower than 64MB then 4 VLFs are created If the growth is between 64MB and 1GB then 8 VLFs are created If the growth is greater than 1GB then 16 VLFs are created. Now the potential for chaos comes if the default values and settings for log file growth are used. By default a database log file gets a 1MB log file with unlimited growth in steps of 10%. The database we just created is 6 MB, let’s add some data and see what happens. USE vlf_test go -- we need somewhere to put the data so, a table is in order IF OBJECT_ID('A_Table') IS NOT NULL DROP TABLE A_Table go CREATE TABLE A_Table ( Col_A int IDENTITY, Col_B CHAR(8000) ) GO -- Let's check the state of the log file -- 4 VLFs found EXECUTE ('DBCC LOGINFO'); go -- We can go ahead and insert some data and then check the state of the log file again INSERT A_Table (col_b) SELECT TOP 500 REPLICATE('a',2000) FROM sys.columns AS sc, sys.columns AS sc2 GO -- insert 500 rows and we get 22 VLFs EXECUTE ('DBCC LOGINFO'); go -- Let's insert more rows INSERT A_Table (col_b) SELECT TOP 2000 REPLICATE('a',2000) FROM sys.columns AS sc, sys.columns AS sc2 GO 10 -- insert 2000 rows, in 10 batches and we suddenly have 107 VLFs EXECUTE ('DBCC LOGINFO'); Well, that escalated quickly! Our log file is split, internally, into 107 fragments after a few thousand inserts. The same happens with any logged transactions, I just chose to illustrate this with INSERTs. Having too many VLFs can cause performance degradation at times of database start up, log backup and log restore operations so it’s well worth keeping a check on this property. How do we prevent excessive VLF creation? Creating the database with larger files and also with larger growth steps and actively choosing to grow your databases rather than leaving it to the Auto Grow event can make sure that the growths are made with a size that is optimal. How do we resolve a situation of a database with too many VLFs? This process needs to be done when the database is under little or no stress so that you don’t affect system users. The steps are: BACKUP LOG YourDBName TO YourBackupDestinationOfChoice Shrink the log file to its smallest possible size DBCC SHRINKFILE(FileNameOfTLogHere, TRUNCATEONLY) * Re-size the log file to the size you want it to, taking in to account your expected needs for the coming months or year. ALTER DATABASE YourDBName MODIFY FILE ( NAME = FileNameOfTLogHere, SIZE = TheSizeYouWantItToBeIn_MB) * – If you don’t know the file name of your log file then run sp_helpfile while you are connected to the database that you want to work on and you will get the details you need. The resize step can take quite a while This is already detailed far better than I can explain it by Kimberley Tripp in her blog 8-Steps-to-better-Transaction-Log-throughput.aspx. The result of this will be a log file with a VLF count according to the bullet list above. Knowing when VLFs are being created By complete coincidence while I have been writing this blog (it’s been quite some time from it’s inception to going live) Jonathan Kehayias from SQLSkills.com has written a great article on how to track database file growth using Event Notifications and Service Broker. I strongly recommend taking a look at it as this is going to catch any sneaky auto grows that take place and let you know about them right away. Hassle free monitoring of VLFs If you are lucky or wise enough to be using SQL Monitor or another monitoring tool that let’s you write your own custom metrics then you can keep an eye on this very easily. There is a custom metric for VLFs (written by Stuart Ainsworth) already on the site and there are some others there are very useful so take a moment or two to look around while you are there. Resources MSDN – http://msdn.microsoft.com/en-us/library/ms179355(v=sql.105).aspx Kimberly Tripp from SQLSkills.com – http://www.sqlskills.com/BLOGS/KIMBERLY/post/8-Steps-to-better-Transaction-Log-throughput.aspx Thomas LaRock at Simple-Talk.com – http://www.simple-talk.com/sql/database-administration/monitoring-sql-server-virtual-log-file-fragmentation/ Disclosure I am a Friend of Red Gate. This means that I am more than likely to say good things about Red Gate DBA and Developer tools. No matter how awesome I make them sound, take the time to compare them with other products before you contact the Red Gate sales team to make your order.

    Read the article

  • SQL University: Database testing and refactoring tools and examples

    - by Mladen Prajdic
    This is a post for a great idea called SQL University started by Jorge Segarra also famously known as SqlChicken on Twitter. It’s a collection of blog posts on different database related topics contributed by several smart people all over the world. So this week is mine and we’ll be talking about database testing and refactoring. In 3 posts we’ll cover: SQLU part 1 - What and why of database testing SQLU part 2 - What and why of database refactoring SQLU part 3 - Database testing and refactoring tools and examples This is the third and last part of the series and in it we’ll take a look at tools we can test and refactor with plus some an example of the both. Tools of the trade First a few thoughts about how to go about testing a database. I'm firmily against any testing tools that go into the database itself or need an extra database. Unit tests for the database and applications using the database should all be in one place using the same technology. By using database specific frameworks we fragment our tests into many places and increase test system complexity. Let’s take a look at some testing tools. 1. NUnit, xUnit, MbUnit All three are .Net testing frameworks meant to unit test .Net application. But we can test databases with them just fine. I use NUnit because I’ve always used it for work and personal projects. One day this might change. So the thing to remember is to be flexible if something better comes along. All three are quite similar and you should be able to switch between them without much problem. 2. TSQLUnit As much as this framework is helpful for the non-C# savvy folks I don’t like it for the reason I stated above. It lives in the database and thus fragments the testing infrastructure. Also it appears that it’s not being actively developed anymore. 3. DbFit I haven’t had the pleasure of trying this tool just yet but it’s on my to-do list. From what I’ve read and heard Gojko Adzic (@gojkoadzic on Twitter) has done a remarkable job with it. 4. Redgate SQL Refactor and Apex SQL Refactor Neither of these refactoring tools are free, however if you have hardcore refactoring planned they are worth while looking into. I’ve only used the Red Gate’s Refactor and was quite impressed with it. 5. Reverting the database state I’ve talked before about ways to revert a database to pre-test state after unit testing. This still holds and I haven’t changed my mind. Also make sure to read the comments as they are quite informative. I especially like the idea of setting up and tearing down the schema for each test group with NHibernate. Testing and refactoring example We’ll take a look at the simple schema and data test for a view and refactoring the SELECT * in that view. We’ll use a single table PhoneNumbers with ID and Phone columns. Then we’ll refactor the Phone column into 3 columns Prefix, Number and Suffix. Lastly we’ll remove the original Phone column. Then we’ll check how the view behaves with tests in NUnit. The comments in code explain the problem so be sure to read them. I’m assuming you know NUnit and C#. T-SQL Code C# test code USE tempdbGOCREATE TABLE PhoneNumbers( ID INT IDENTITY(1,1), Phone VARCHAR(20))GOINSERT INTO PhoneNumbers(Phone)SELECT '111 222333 444' UNION ALLSELECT '555 666777 888'GO-- notice we don't have WITH SCHEMABINDINGCREATE VIEW vPhoneNumbersAS SELECT * FROM PhoneNumbersGO-- Let's take a look at what the view returns -- If we add a new columns and rows both tests will failSELECT *FROM vPhoneNumbers GO -- DoesViewReturnCorrectColumns test will SUCCEED -- DoesViewReturnCorrectData test will SUCCEED -- refactor to split Phone column into 3 partsALTER TABLE PhoneNumbers ADD Prefix VARCHAR(3)ALTER TABLE PhoneNumbers ADD Number VARCHAR(6)ALTER TABLE PhoneNumbers ADD Suffix VARCHAR(3)GO-- update the new columnsUPDATE PhoneNumbers SET Prefix = LEFT(Phone, 3), Number = SUBSTRING(Phone, 5, 6), Suffix = RIGHT(Phone, 3)GO-- remove the old columnALTER TABLE PhoneNumbers DROP COLUMN PhoneGO-- This returns unexpected results!-- it returns 2 columns ID and Phone even though -- we don't have a Phone column anymore.-- Notice that the data is from the Prefix column-- This is a danger of SELECT *SELECT *FROM vPhoneNumbers -- DoesViewReturnCorrectColumns test will SUCCEED -- DoesViewReturnCorrectData test will FAIL -- for a fix we have to call sp_refreshview -- to refresh the view definitionEXEC sp_refreshview 'vPhoneNumbers'-- after the refresh the view returns 4 columns-- this breaks the input/output behavior of the database-- which refactoring MUST NOT doSELECT *FROM vPhoneNumbers -- DoesViewReturnCorrectColumns test will FAIL -- DoesViewReturnCorrectData test will FAIL -- to fix the input/output behavior change problem -- we have to concat the 3 columns into one named PhoneALTER VIEW vPhoneNumbersASSELECT ID, Prefix + ' ' + Number + ' ' + Suffix AS PhoneFROM PhoneNumbersGO-- now it works as expectedSELECT *FROM vPhoneNumbers -- DoesViewReturnCorrectColumns test will SUCCEED -- DoesViewReturnCorrectData test will SUCCEED -- clean upDROP VIEW vPhoneNumbersDROP TABLE PhoneNumbers [Test]public void DoesViewReturnCoorectColumns(){ // conn is a valid SqlConnection to the server's tempdb // note the SET FMTONLY ON with which we return only schema and no data using (SqlCommand cmd = new SqlCommand("SET FMTONLY ON; SELECT * FROM vPhoneNumbers", conn)) { DataTable dt = new DataTable(); dt.Load(cmd.ExecuteReader(CommandBehavior.CloseConnection)); // test returned schema: number of columns, column names and data types Assert.AreEqual(dt.Columns.Count, 2); Assert.AreEqual(dt.Columns[0].Caption, "ID"); Assert.AreEqual(dt.Columns[0].DataType, typeof(int)); Assert.AreEqual(dt.Columns[1].Caption, "Phone"); Assert.AreEqual(dt.Columns[1].DataType, typeof(string)); }} [Test]public void DoesViewReturnCorrectData(){ // conn is a valid SqlConnection to the server's tempdb using (SqlCommand cmd = new SqlCommand("SELECT * FROM vPhoneNumbers", conn)) { DataTable dt = new DataTable(); dt.Load(cmd.ExecuteReader(CommandBehavior.CloseConnection)); // test returned data: number of rows and their values Assert.AreEqual(dt.Rows.Count, 2); Assert.AreEqual(dt.Rows[0]["ID"], 1); Assert.AreEqual(dt.Rows[0]["Phone"], "111 222333 444"); Assert.AreEqual(dt.Rows[1]["ID"], 2); Assert.AreEqual(dt.Rows[1]["Phone"], "555 666777 888"); }}   With this simple example we’ve seen how a very simple schema can cause a lot of problems in the whole application/database system if it doesn’t have tests. Imagine what would happen if some outside process would depend on that view. It would get wrong data and propagate it silently throughout the system. And that is not good. So have tests at least for the crucial parts of your systems. And with that we conclude the Database Testing and Refactoring week at SQL University. Hope you learned something new and enjoy the learning weeks to come. Have fun!

    Read the article

  • SQL SERVER – SSMS: Database Consistency History Report

    - by Pinal Dave
    Doctor and Database The last place I like to visit is always a hospital. With the monsoon season starting, intermittent rains, it has become sort of a routine to get a cycle of fever every other year (seriously I hate it). So when I visit my doctor, it is always interesting in the way he quizzes me. The routine question of – “How many days have you had this?”, “Is there any pattern?”, “Did you drench in rain?”, “Do you have any other symptom?” and so on. The idea here is that the doctor wants to find any anomaly or a pattern that will guide him to a viral or bacterial type. Most of the time they get it based on experience and sometimes after a battery of tests. So if there is consistent behavior to your problem, there is always a solution out. SQL Server has its way to find if the server data / files are in consistent state using the DBCC commands. Back to SQL Server In real life, Database consistency check is one of the critical operations a DBA generally doesn’t give much priority. Many readers of my blogs have asked many times, how do we know if the database is consistent? How do I read output of DBCC CHECKDB and find if everything is right or not? My common answer to all of them is – look at the bottom of checkdb (or checktable) output and look for below line. CHECKDB found 0 allocation errors and 0 consistency errors in database ‘DatabaseName’. Above is a “good sign” because we are seeing zero allocation and zero consistency error. If you are seeing non-zero errors then there is some problem with the database. Sample output is shown as below: CHECKDB found 0 allocation errors and 2 consistency errors in database ‘DatabaseName’. repair_allow_data_loss is the minimum repair level for the errors found by DBCC CHECKDB (DatabaseName). If we see non-zero error then most of the time (not always) we get repair options depending on the level of corruption. There is risk involved with above option (repair_allow_data_loss), that is – we would lose the data. Sometimes the option would be repair_rebuild which is little safer. Though these options are available, it is important to find the root cause to the problem. In standard report, there is a report which can show the history of checkdb executed for the selected database. Since this is a database level report, we need to right click on database, click Reports, click Standard Reports and then choose “Database Consistency History” report. The information in this report is picked from default trace. If default trace is disabled or there is no checkdb run or information is not there in default trace (because it’s rolled over), we would get report like below. As we can see report says it very clearly: Currently, no execution history of CHECKDB is available or default trace is not enabled. To demonstrate, I have caused corruption in one of the database and did below steps. Run CheckDB so that errors are reported. Fix the corruption by losing the data using repair option Run CheckDB again to check if corruption is cleared. After that I have launched the report and below is what we would see. If you are lazy like me and don’t want to run the report manually for each database then below query would be handy to provide same report for all database. This query is runs behind the scenes by the report. All I have done is remove the filter for database name (at the last – highlighted). DECLARE @curr_tracefilename VARCHAR(500); DECLARE @base_tracefilename VARCHAR(500); DECLARE @indx INT; SELECT @curr_tracefilename = path FROM sys.traces WHERE is_default = 1; SET @curr_tracefilename = REVERSE(@curr_tracefilename); SELECT @indx  = PATINDEX('%\%', @curr_tracefilename) ; SET @curr_tracefilename = REVERSE(@curr_tracefilename); SET @base_tracefilename = LEFT( @curr_tracefilename,LEN(@curr_tracefilename) - @indx) + '\log.trc'; SELECT  SUBSTRING(CONVERT(NVARCHAR(MAX),TEXTData),36, PATINDEX('%executed%',TEXTData)-36) AS command ,       LoginName ,       StartTime ,       CONVERT(INT,SUBSTRING(CONVERT(NVARCHAR(MAX),TEXTData),PATINDEX('%found%',TEXTData) +6,PATINDEX('%errors %',TEXTData)-PATINDEX('%found%',TEXTData)-6)) AS errors ,       CONVERT(INT,SUBSTRING(CONVERT(NVARCHAR(MAX),TEXTData),PATINDEX('%repaired%',TEXTData) +9,PATINDEX('%errors.%',TEXTData)-PATINDEX('%repaired%',TEXTData)-9)) repaired ,       SUBSTRING(CONVERT(NVARCHAR(MAX),TEXTData),PATINDEX('%time:%',TEXTData)+6,PATINDEX('%hours%',TEXTData)-PATINDEX('%time:%',TEXTData)-6)+':'+SUBSTRING(CONVERT(NVARCHAR(MAX),TEXTData),PATINDEX('%hours%',TEXTData) +6,PATINDEX('%minutes%',TEXTData)-PATINDEX('%hours%',TEXTData)-6)+':'+SUBSTRING(CONVERT(NVARCHAR(MAX),TEXTData),PATINDEX('%minutes%',TEXTData) +8,PATINDEX('%seconds.%',TEXTData)-PATINDEX('%minutes%',TEXTData)-8) AS time FROM::fn_trace_gettable( @base_tracefilename, DEFAULT) WHERE EventClass = 22 AND SUBSTRING(TEXTData,36,12) = 'DBCC CHECKDB' -- AND DatabaseName = @DatabaseName; Don’t get worried about the logic above. All it is doing is reading the trace files, parsing below entry and getting out information for underlined words. DBCC CHECKDB (CorruptedDatabase) executed by sa found 2 errors and repaired 0 errors. Elapsed time: 0 hours 0 minutes 0 seconds.  Internal database snapshot has split point LSN = 00000029:00000030:0001 and first LSN = 00000029:00000020:0001. Hopefully now onwards you would run checkdb and understand the importance of it. As responsible DBAs I am sure you are already doing it, let me know how often do you actually run them on you production environment? Reference: Pinal Dave (http://blog.sqlauthority.com)Filed under: PostADay, SQL, SQL Authority, SQL Query, SQL Server, SQL Server Management Studio, SQL Tips and Tricks, T SQL Tagged: SQL Reports

    Read the article

  • Where can I find accessible bug/issue databases with complete revision history

    - by namenlos
    I'm performing some research and analysis on bug/issue tracking databases and more specifically on how programmers and teams of programmers actually interact with them. What I'm looking for involves understanding how those databases change over time. So what I don't need for example: is a database of all the bugs of some open source project as the bugs exist today. What I do need is a complete set of revision history for every issue/bug in the database. This would enable me to pick a specific datetime and say here were the list of all the issues/bugs that existed at that moment in time. Anyway know of some publicly accessible issue/bug databases that expose this revision data? Ideally, the revision would look something like this (shown for a single bug, with two revisions) ISSUEID PRI SEV ASSIGNEDTO MODIFIEDON VALIDUNTIL 1 2 2 mel apr-1-2010:5pm apr-1-2010:6pm 1 2 3 steve apr-1-2010:6pm NULL

    Read the article

  • Local LINQtoSQL Database For Your Windows Phone 7 Application

    - by Tim Murphy
    There aren’t many applications that are of value without having some for of data store.  In Windows Phone development we have a few options.  You can store text directly to isolated storage.  You can also use a number of third party libraries to create or mimic databases in isolated storage.  With Mango we gained the ability to have a native .NET database approach which uses LINQ to SQL.  In this article I will try to bring together the components needed to implement this last type of data store and fill in some of the blanks that I think other articles have left out. Defining A Database The first things you are going to need to do is define classes that represent your tables and a data context class that is used as the overall database definition.  The table class consists of column definitions as you would expect.  They can have relationships and constraints as with any relational DBMS.  Below is an example of a table definition. First you will need to add some assembly references to the code file. using System.ComponentModel;using System.Data.Linq;using System.Data.Linq.Mapping; You can then add the table class and its associated columns.  It needs to implement INotifyPropertyChanged and INotifyPropertyChanging.  Each level of the class needs to be decorated with the attribute appropriate for that part of the definition.  Where the class represents the table the properties represent the columns.  In this example you will see that the column is marked as a primary key and not nullable with a an auto generated value. You will also notice that the in the column property’s set method It uses the NotifyPropertyChanging and NotifyPropertyChanged methods in order to make sure that the proper events are fired. [Table]public class MyTable: INotifyPropertyChanged, INotifyPropertyChanging{ public event PropertyChangedEventHandler PropertyChanged; private void NotifyPropertyChanged(string propertyName) { if(PropertyChanged != null) { PropertyChanged(this, new PropertyChangedEventArgs(propertyName)); } } public event PropertyChangingEventHandler PropertyChanging; private void NotifyPropertyChanging(string propertyName) { if(PropertyChanging != null) { PropertyChanging(this, new PropertyChangingEventArgs(propertyName)); } } private int _TableKey; [Column(IsPrimaryKey = true, IsDbGenerated = true, DbType = "INT NOT NULL Identity", CanBeNull = false, AutoSync = AutoSync.OnInsert)] public int TableKey { get { return _TableKey; } set { NotifyPropertyChanging("TableKey"); _TableKey = value; NotifyPropertyChanged("TableKey"); } } The last part of the database definition that needs to be created is the data context.  This is a simple class that takes an isolated storage location connection string its constructor and then instantiates tables as public properties. public class MyDataContext: DataContext{ public MyDataContext(string connectionString): base(connectionString) { MyRecords = this.GetTable<MyTable>(); } public Table<MyTable> MyRecords;} Creating A New Database Instance Now that we have a database definition it is time to create an instance of the data context within our Windows Phone app.  When your app fires up it should check if the database already exists and create an instance if it does not.  I would suggest that this be part of the constructor of your ViewModel. db = new MyDataContext(connectionString);if(!db.DatabaseExists()){ db.CreateDatabase();} The next thing you have to know is how the connection string for isolated storage should be constructed.  The main sticking point I have found is that the database cannot be created unless the file mode is read/write.  You may have different connection strings but the initial one needs to be similar to the following. string connString = "Data Source = 'isostore:/MyApp.sdf'; File Mode = read write"; Using you database Now that you have done all the up front work it is time to put the database to use.  To make your life a little easier and keep proper separation between your view and your viewmodel you should add a couple of methods to the viewmodel.  These will do the CRUD work of your application.  What you will notice is that the SubmitChanges method is the secret sauce in all of the methods that change data. private myDataContext myDb;private ObservableCollection<MyTable> _viewRecords;public ObservableCollection<MyTable> ViewRecords{ get { return _viewRecords; } set { _viewRecords = value; NotifyPropertyChanged("ViewRecords"); }}public void LoadMedstarDbData(){ var tempItems = from MyTable myRecord in myDb.LocalScans select myRecord; ViewRecords = new ObservableCollection<MyTable>(tempItems);}public void SaveChangesToDb(){ myDb.SubmitChanges();}public void AddMyTableItem(MyTable newScan){ myDb.LocalScans.InsertOnSubmit(newScan); myDb.SubmitChanges();}public void DeleteMyTableItem(MyTable newScan){ myDb.LocalScans.DeleteOnSubmit(newScan); myDb.SubmitChanges();} Updating existing database What happens when you need to change the structure of your database?  Unfortunately you have to add code to your application that checks the version of the database which over time will create some pollution in your codes base.  On the other hand it does give you control of the update.  In this example you will see the DatabaseSchemaUpdater in action.  Assuming we added a “Notes” field to the MyTable structure, the following code will check if the database is the latest version and add the field if it isn’t. if(!myDb.DatabaseExists()){ myDb.CreateDatabase();}else{ DatabaseSchemaUpdater dbUdater = myDb.CreateDatabaseSchemaUpdater(); if(dbUdater.DatabaseSchemaVersion < 2) { dbUdater.AddColumn<MyTable>("Notes"); dbUdater.DatabaseSchemaVersion = 2; dbUdater.Execute(); }} Summary This approach does take a fairly large amount of work, but I think the end product is robust and very native for .NET developers.  It turns out to be worth the investment. del.icio.us Tags: Windows Phone,Windows Phone 7,LINQ to SQL,LINQ,Database,Isolated Storage

    Read the article

  • Advice on software / database design to avoid using cursors when updating database

    - by Remnant
    I have a database that logs when an employee has attended a course and when they are next due to attend the course (courses tend to be annual). As an example, the following employee attended course '1' on 1st Jan 2010 and, as the course is annual, is due to attend next on the 1st Jan 2011. As today is 20th May 2010 the course status reads as 'Complete' i.e. they have done the course and do not need to do it again until next year: EmployeeID CourseID AttendanceDate DueDate Status 123456 1 01/01/2010 01/01/2011 Complete In terms of the DueDate I calculate this in SQL when I update the employee's record e.g. DueDate = AttendanceDate + CourseFrequency (I pull course frequency this from a separate table). In my web based app (asp.net mvc) I pull back this data for all employees and display it in a grid like format for HR managers to review. This allows HR to work out who needs to go on courses. The issue I have is as follows. Taking the example above, suppose today is 2nd Jan 2011. In this case, employee 123456 is now overdue for the course and I would like to set the Status to Incomplete so that the HR manager can see that they need to action this i.e. get employee on the course. I could build a trigger in the database to run overnight to update the Status field for all employees based on the current date. From what I have read I would need to use cursors to loop over each row to amend the status and this is considered bad practice / inefficient or at least something to avoid if you can??? Alternatively, I could compute the Status in my C# code after I have pulled back the data from the database and before I display it on screen. The issue with this is that the Status in the database would not necessarily match what is shown on screen which just feels plain wrong to me. Does anybody have any advice on the best practice approach to such an issue? It helps, if I did use a cursor I doubt I would be looping over more than 1000 records at any given time. Maybe this is such small volume that using cursors is okay?

    Read the article

  • Better way to design a database

    - by cMinor
    I have a conceptual problem and I would like to get your ideas on how I'll be able to do what I am aiming. My goal is to create a database with information of persons who work at a place depending on their profession and skills,and keep control of salary and projects (how much would cost summing all the hours of work) I have 3 categories which can have subcategories: Outsourcing Technician welder turner assistant Administrative supervisor manager So each person has its information and the projects they are working on, also one person may do several jobs... I was thinking about having 5 tables (EMPLOYEE, SKILLS, PROYECTS, SALARY, PROFESSION) but I guess there is a better way of doing this. create table Employee ( PRIMARY KEY [Person_ID] int(10), [Name] varchar(30), [sex] varchar(10), [address] varchar(10), [profession] varchar(10), [Skills_ID] int(10), [Proyect_ID] int(10), [Salary_ID] int(10), [Salary] float ) create table Skills ( PRIMARY KEY [Skills_ID] int(10), FOREIGN KEY [Skills_name] varchar(10) REFERENCES Employee(Person_ID), [Skills_pay] float(10), [Comments] varchar(50) ) create table Proyects ( PRIMARY KEY [Proyect_ID] int(10), FOREIGN KEY [Skills_name] varchar(10) REFERENCES Employee(Person_ID) [Proyect_name] varchar(10), [working_Hours] float(10), [Comments] varchar(50) ) create table Salary ( PRIMARY KEY [Salary_ID] int(10), FOREIGN KEY [Skills_name] varchar(10) REFERENCES Employee(Person_ID) [Proyect_name] varchar(10), [working_Hours] float(10), [Comments] varchar(50) ) So to get the total amount of the cost of a project I would just sum the working hours of each employee envolved and sum some extra costs in an aggregate query. Is there a way to do this in a more efficient way? What to add or delete of this small model? I guess I am missing something in the salary - maybe I need another table for that?

    Read the article

  • Oracle Database Recovery Problem

    - by Palani
    I am very new to Oracle, and trying to restore a oracle 8i database on win 2000 server. I have one week old database backup (backup taken with exp command), and i want to restore it now. Now I am unable to login through sqlplus (got shutdown in progress error) I have a backup and i want to restore it, but oracle is not starting at all, and 'imp' command is failing. I started sqlplus / as sysdba and following is the log of what i am trying to do. Can some one guide me further. SQL> shutdown immediate; ORA-01109: database not open Database dismounted. ORACLE instance shut down. SQL> startup; ORACLE instance started. Total System Global Area 143423516 bytes Fixed Size 75804 bytes Variable Size 58105856 bytes Database Buffers 85164032 bytes Redo Buffers 77824 bytes Database mounted. ORA-01589: must use RESETLOGS or NORESETLOGS option for database open SQL> shutdown immediate; ORA-01109: database not open Database dismounted. ORACLE instance shut down. SQL> startup mount; ORACLE instance started. Total System Global Area 143423516 bytes Fixed Size 75804 bytes Variable Size 58105856 bytes Database Buffers 85164032 bytes Redo Buffers 77824 bytes Database mounted. SQL> alter database open; alter database open * ERROR at line 1: ORA-01589: must use RESETLOGS or NORESETLOGS option for database open SQL> alter database open resetlogs; alter database open resetlogs * ERROR at line 1: ORA-01245: offline file 1 will be lost if RESETLOGS is done ORA-01110: data file 1: 'C:\ORACLE\ORADATA\ABCD\SYSTEM01.DBF'

    Read the article

  • starting oracle database automatically.

    - by Searock
    I am using Fedora 8 and Oracle 10g Express Edition. Every time I start my fedora I have to click on start database. How can I add startdb.sh to startup so that it automatically executes when Fedora starts? I have tried adding the path to /etc/rc.d/rc.local but it still doesn't work. ./usr/lib/oracle/xe/app/oracle/product/10.2.0/server/config/scripts/startdb.sh I have even tried to add this script in /etc/init.d/oracle #!/bin/bash # # Run-level Startup script for the Oracle Instance and Listener # # chkconfig: 345 91 19 # description: Startup/Shutdown Oracle listener and instance ORA_HOME="/u01/app/oracle/product/9.2.0.1.0" ORA_OWNR="oracle" # if the executables do not exist -- display error if [ ! -f $ORA_HOME/bin/dbstart -o ! -d $ORA_HOME ] then echo "Oracle startup: cannot start" exit 1 fi # depending on parameter -- startup, shutdown, restart # of the instance and listener or usage display case "$1" in start) # Oracle listener and instance startup echo -n "Starting Oracle: " su - $ORA_OWNR -c "$ORA_HOME/bin/lsnrctl start" su - $ORA_OWNR -c $ORA_HOME/bin/dbstart touch /var/lock/subsys/oracle echo "OK" ;; stop) # Oracle listener and instance shutdown echo -n "Shutdown Oracle: " su - $ORA_OWNR -c "$ORA_HOME/bin/lsnrctl stop" su - $ORA_OWNR -c $ORA_HOME/bin/dbshut rm -f /var/lock/subsys/oracle echo "OK" ;; reload|restart) $0 stop $0 start ;; *) echo "Usage: $0 start|stop|restart|reload" exit 1 esac exit 0 and even this doesn't work. startdb.sh is located at /usr/lib/oracle/xe/app/oracle/product/10.2.0/server/config/scripts/startdb.sh Thanks.

    Read the article

  • Oracle Database Recovery Problem

    - by Palani
    I am very new to Oracle, and trying to restore a oracle 8i database on win 2000 server. I have one week old database backup (backup taken with exp command), and i want to restore it now. Now I am unable to login through sqlplus (got shutdown in progress error) I have a backup and i want to restore it, but oracle is not starting at all, and 'imp' command is failing. I started sqlplus / as sysdba and following is the log of what i am trying to do. Can some one guide me further. SQL> shutdown immediate; ORA-01109: database not open Database dismounted. ORACLE instance shut down. SQL> startup; ORACLE instance started. Total System Global Area 143423516 bytes Fixed Size 75804 bytes Variable Size 58105856 bytes Database Buffers 85164032 bytes Redo Buffers 77824 bytes Database mounted. ORA-01589: must use RESETLOGS or NORESETLOGS option for database open SQL> shutdown immediate; ORA-01109: database not open Database dismounted. ORACLE instance shut down. SQL> startup mount; ORACLE instance started. Total System Global Area 143423516 bytes Fixed Size 75804 bytes Variable Size 58105856 bytes Database Buffers 85164032 bytes Redo Buffers 77824 bytes Database mounted. SQL> alter database open; alter database open * ERROR at line 1: ORA-01589: must use RESETLOGS or NORESETLOGS option for database open SQL> alter database open resetlogs; alter database open resetlogs * ERROR at line 1: ORA-01245: offline file 1 will be lost if RESETLOGS is done ORA-01110: data file 1: 'C:\ORACLE\ORADATA\ABCD\SYSTEM01.DBF'

    Read the article

  • Tools to manage sql 2008 database mirroring?

    - by lemkepf
    We are going to be moving about 20 databases that live on a single instance of sql 2000 to a sql 2008 r2 environment with database mirroring. What I'm looking for is a tool or scripts that will help me manage the conversion and management of those 20db's onto this new mirrored environment easily. There are many steps in setting each DB up and I want to automate as much as possible. Edit: Here are the steps I've been doing manually: Create the same username/passwords from the old sql 2000 server onto new sql 2008 server. Then sync those users/passwords onto the other sql 2008 server with the same SSID's so when we do the db backup and restore they match up. Take a backup of each sql 2000 db's. Copy them to server A. Restore the backup to server A. Backup from server a, copy to server b, restore there. Run the mirror "configure security" wizard. Start mirroring. I've love to be able to script this out or have a tool that does it for me. Thanks! Paul

    Read the article

  • Delphi: how to create Firebird database programmatically

    - by Brad
    I'm using D2K9, Zeos 7Alpha, and Firebird 2.1 I had this working before I added the autoinc field. Although I'm not sure I was doing it 100% correctly. I don' know what order to do the SQL code, with the triggers, Generators, etc.. I've tried several combinations, I'm guessing I'm doing something wrong other than just that for this not to work. SQL File From IB Expert : /********************************************/ /* Generated by IBExpert 5/4/2010 3:59:48 PM / /*********************************************/ /********************************************/ /* Following SET SQL DIALECT is just for the Database Comparer / /*********************************************/ SET SQL DIALECT 3; /********************************************/ /* Tables / /*********************************************/ CREATE GENERATOR GEN_EMAIL_ACCOUNTS_ID; CREATE TABLE EMAIL_ACCOUNTS ( ID INTEGER NOT NULL, FNAME VARCHAR(35), LNAME VARCHAR(35), ADDRESS VARCHAR(100), CITY VARCHAR(35), STATE VARCHAR(35), ZIPCODE VARCHAR(20), BDAY DATE, PHONE VARCHAR(20), UNAME VARCHAR(255), PASS VARCHAR(20), EMAIL VARCHAR(255), CREATEDDATE DATE, "ACTIVE" BOOLEAN DEFAULT 0 NOT NULL /* BOOLEAN = SMALLINT CHECK (value is null or value in (0, 1)) /, BANNED BOOLEAN DEFAULT 0 NOT NULL / BOOLEAN = SMALLINT CHECK (value is null or value in (0, 1)) /, "PUBLIC" BOOLEAN DEFAULT 0 NOT NULL / BOOLEAN = SMALLINT CHECK (value is null or value in (0, 1)) */, NOTES BLOB SUB_TYPE 0 SEGMENT SIZE 1024 ); /********************************************/ /* Primary Keys / /*********************************************/ ALTER TABLE EMAIL_ACCOUNTS ADD PRIMARY KEY (ID); /********************************************/ /* Triggers / /*********************************************/ SET TERM ^ ; /********************************************/ /* Triggers for tables / /*********************************************/ /* Trigger: EMAIL_ACCOUNTS_BI */ CREATE OR ALTER TRIGGER EMAIL_ACCOUNTS_BI FOR EMAIL_ACCOUNTS ACTIVE BEFORE INSERT POSITION 0 AS BEGIN IF (NEW.ID IS NULL) THEN NEW.ID = GEN_ID(GEN_EMAIL_ACCOUNTS_ID,1); END ^ SET TERM ; ^ /********************************************/ /* Privileges / /*********************************************/ Triggers: /********************************************/ /* Following SET SQL DIALECT is just for the Database Comparer / /*********************************************/ SET SQL DIALECT 3; CREATE GENERATOR GEN_EMAIL_ACCOUNTS_ID; SET TERM ^ ; CREATE OR ALTER TRIGGER EMAIL_ACCOUNTS_BI FOR EMAIL_ACCOUNTS ACTIVE BEFORE INSERT POSITION 0 AS BEGIN IF (NEW.ID IS NULL) THEN NEW.ID = GEN_ID(GEN_EMAIL_ACCOUNTS_ID,1); END ^ SET TERM ; ^ Generators: CREATE SEQUENCE GEN_EMAIL_ACCOUNTS_ID; ALTER SEQUENCE GEN_EMAIL_ACCOUNTS_ID RESTART WITH 2; /* Old syntax is: CREATE GENERATOR GEN_EMAIL_ACCOUNTS_ID; SET GENERATOR GEN_EMAIL_ACCOUNTS_ID TO 2; */ My Code: procedure TForm2.New1Click(Sender: TObject); var query:string; begin if JvOpenDialog1.Execute then begin ZConnection1.Disconnect; ZConnection1.Database:= jvOpenDialog1.FileName; if not FileExists(ZConnection1.database) then begin ZConnection1.Properties.Add('createnewdatabase=create database '''+ZConnection1.Database+''' user ''sysdba'' password ''masterkey'' page_size 4096 default character set iso8859_2;'); try ZConnection1.Connect; except ShowMessage('Error Connection To Database File'); application.Terminate; end; end else begin ShowMessage('Database File Already Exists.'); exit; end; end; query := 'CREATE DOMAIN BOOLEAN AS SMALLINT CHECK (value is null or value in (0, 1))'; Zconnection1.ExecuteDirect(query); query:='CREATE TABLE EMAIL_ACCOUNTS (ID INTEGER NOT NULL,FNAME VARCHAR(35),LNAME VARCHAR(35),'+ 'ADDRESS VARCHAR(100), CITY VARCHAR(35), STATE VARCHAR(35), ZIPCODE VARCHAR(20),' + 'BDAY DATE, PHONE VARCHAR(20), UNAME VARCHAR(255), PASS VARCHAR(20),' + 'EMAIL VARCHAR(255),CREATEDDATE DATE , '+ '"ACTIVE" BOOLEAN DEFAULT 0 NOT NULL,'+ 'BANNED BOOLEAN DEFAULT 0 NOT NULL,'+ '"PUBLIC" BOOLEAN DEFAULT 0 NOT NULL,' + 'NOTES BLOB SUB_TYPE 0 SEGMENT SIZE 1024)'; //ZConnection.ExecuteDirect('CREATE TABLE NOTES (noteTitle TEXT PRIMARY KEY,noteDate DATE,noteNote TEXT)'); Zconnection1.ExecuteDirect(query); { } query := 'CREATE SEQUENCE GEN_EMAIL_ACCOUNTS_ID;'+ 'ALTER SEQUENCE GEN_EMAIL_ACCOUNTS_ID RESTART WITH 1'; Zconnection1.ExecuteDirect(query); query := 'ALTER TABLE EMAIL_ACCOUNTS ADD PRIMARY KEY (ID)'; Zconnection1.ExecuteDirect(query); query := 'SET TERM ^'; Zconnection1.ExecuteDirect(query); query := 'CREATE OR ALTER TRIGGER EMAIL_ACCOUNTS_BI FOR EMAIL_ACCOUNTS'+ 'ACTIVE BEFORE INSERT POSITION 0'+ 'AS'+ 'BEGIN'+ 'IF (NEW.ID IS NULL) THEN'+ 'NEW.ID = GEN_ID(GEN_EMAIL_ACCOUNTS_ID,1);'+ 'END'+ '^'+ 'SET TERM ; ^'; Zconnection1.ExecuteDirect(query); ZTable1.Active:=true; end;

    Read the article

  • My father is a doctor. He is insisting on writing a database to store non-critical patient information, with no programming background

    - by Dominic Bou-Samra
    So, my father is currently in the process of "hacking" together a database using FileMaker Pro, a GUI based databasing tool for his small (4 doctor) practice. The database will be used to help ease the burden on reporting from medical machines, streamlining quite a clumsy process. He's got no programming background, and seems to be doing everything in his power to not learn things correctly. He's got duplicate data types, no database-enforced relationships (foreign/primary key constraints) and a dozen other issues. He's doing it all by hand via GUI tool using Youtube videos. My issue is, that whilst I want him to succeed 100%, I don't think it's appropriate for him to be handling these types of decisions. How do I convince him that without some sort of education in these topics, a hacked together solution is a bad idea? He's can be quite stubborn and I think he sees these types of jobs as "childs play" How should I approach this? Is it even that bad an idea - or am I correct in thinking he should hire a proper DBA/developer to handle this so that it doesn't become a maintenance nightmare? NB: I am a developer consultant of 4 years and I've seen my share of painful customer implementations.

    Read the article

  • PaaS, DBaaS and the Oracle Database Cloud Service

    - by yaldahhakim
    v\:* {behavior:url(#default#VML);} o\:* {behavior:url(#default#VML);} w\:* {behavior:url(#default#VML);} .shape {behavior:url(#default#VML);} Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:"Times New Roman"; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi;} As with many widely hyped areas, there is much more variation within the broad spectrum of products referred to as “Cloud” that is immediately apparent. This variation is evident in one of the key misunderstandings about the Oracle Database Cloud Service. People could be forgiven for thinking that the Database Cloud Service was a Database-as-a-Service (DBaaS), but this is actually not true. The Database Cloud Service is a Platform-as-a-Service, which presents a different user and developer interface and has a different set of qualities. A good way to think about the difference between these two varieties of Cloud offerings is that you, the customer, have to deal with things at the level of the offering, but not for anything below it. In practice, this means that you do not have to deal with hardware or system software, including installation and maintenance, for DBaaS. You also do not have much control over configuration of these options. For PaaS, you don’t have to deal with hardware, system software, or database software – and also do not have control over these levels in the stack. So you cannot modify configuration parameters for the database with the Database Cloud Service – your interface is through SQL and PL/SQL, with Application Express, included in the Database Cloud Service, or through JDBC for Java apps running in the Java Cloud Service, or through RESTful Web Services. You will notice what is not mentioned there – SQL*Net. You cannot access your Oracle Database Cloud Service by changing an entry in the TNSNames file and using SQL*Net. So the effort involved in migrating an existing Oracle Database in your data center to the Database Cloud Service may be prohibitive. The good news is that Application Express and the RESTful Web Services wizard in the Database Cloud Service allow you to develop new applications very quickly, and, of course, the provisioning of the entire Database Cloud Service takes only minutes.

    Read the article

  • How should I structure my database to gain maximum efficiently in this scenario?

    - by Bob Jansen
    I'm developing a PHP script that analyzes the web traffic of my clients websites. By placing a link to a javascript on the clients website (think of Google Analyses), my script harvests information like: the visitors IP address, reference link, current page link, user agent, etc. Now my clients can view these statistics via a control panel that I have build. These clients can also adjust profile settings, set firewall rules, create support tickets and pay invoices. Currently all the the traffic is stored in one table. You can imagine that this tabel would become very large as some my clients receive thousands of pageviews per day. Furthermore, all the traffic data of each client would be stored in the same table, creating a mess. This is the same for the firewall rules currently, and the invoice and support system. I'm looking for way to structure my database in a more organized way to hold large amounts of data of multiple users. This is the first project that I'm developing that deals with so much data, and would like to hear suggestions and tips. I was thinking of using multiple databases to structure the data. The main database will store users data (email,pass,id,etc) admin/website settings. Than each client will have an unique database labeled prefix_userid, which carry tables holding their traffic, invoice, and support ticket data. Would this be a solution, and would it slow down or speed up overall performances (that is spreading the data over muliple databases). I have a solid VPS, but would like to safe and be as effient as possible.

    Read the article

  • Oracle SQL Developer is for Oracle Database

    - by thatjeffsmith
    What is Oracle SQL Developer? Well, according to this document on OTN… What is SQL Developer? Date: May 2014 Oracle SQL Developer is the Oracle Database IDE. A free graphical user interface, Oracle SQL Developer allows database users and administrators to do their database tasks in fewer clicks and keystrokes. A productivity tool, SQL Developer’s main objective is to help the end user save time and maximize the return on investment in the Oracle Database technology stack. Ok, sounds pretty straightforward. Where does the confusion lie then? Some People Use SQL Developer to Connect to 3rd Party Databases SQL Developer allows you to register 3rd party database JDBC drivers. The 3rd party being a company OTHER than Oracle that makes a database product. You know who they are (SAP, MSFT, IBM, etc.) Registering 3rd party JDBC drivers in SQL Developer But maybe you don’t understand why we support these types of connections? It’s for one driving reason. To Help You Migrate to Oracle Database Yes, you get a worksheet and a tree to query and browse those systems. But, the real meat and bones there are around our migration projects and our translation scratch editor. At the end of the day, it’s there so you can move your data from say Sybase ASE to Oracle Database. On a side note, the migration technology was previously available in a separate application, the Migration Workbench. The technology and the awesome people behind it were folded into SQL Developer. So when asked what SQL Developer is, I say it’s the Database IDE and the official 3rd party database migration to Oracle platform. So anyways, when you ask for better support for another 3rd party provider, we deliver that support based on that business driver. If another 3rd party database jdbc driver is introduced, it’s because we have a lot of customers migrating from that platform. We’re not adding it to make it easier for you to work with SQL Server on your Mac. But, if you find that useful – that is cool. It’s just not why we’ve got the support for SQL Server connections in SQL Developer.

    Read the article

  • SQL SERVER – How to easily work with Database Diagrams

    - by Pinal Dave
    Databases are very widely used in the modern world. Regardless of the complexity of a database, each one requires in depth designing. To practice along please Download dbForge Studio now.  The right methodology of designing a database is based on the foundations of data normalization, according to which we should first define database’s key elements – entities. Afterwards the attributes of entities and relations between them are determined. There is a strong opinion that the process of database designing should start with a pencil and a blank sheet of paper. This might look old-fashioned nowadays, because SQL Server provides a much wider functionality for designing databases – Database Diagrams. When using SSMS for working with Database Diagrams I realized two things – on the one hand, visualization of a scheme allows designing a database more efficiently; on the other – when it came to creating a big scheme, some difficulties occurred when designing with SSMS. The alternatives haven’t taken long to wait and dbForge Studio for SQL Server is one of them. Its functions offer more advantages for working with Database Diagrams. For example, unlike SSMS, dbForge Studio supports an opportunity to drag-and-drop several tables at once from the Database Explorer. This is my opinion but personally I find this option very useful. Another great thing is that a diagram can be saved as both a graphic file and a special XML file, which in case of identical environment can be easily opened on the other server for continuing the work. During working with dbForge Studio it turned out that it offers a wide set of elements to operate with on the diagram. Noteworthy among such elements are containers which allow aggregating diagram objects into thematic groups. Moreover, you can even place an image directly on the diagram if the scheme design is based on a standard template. Each of the development environments has a different approach to storing a diagram (for example, SSMS stores them on a server-side, whereas dbForge Studio – in a local file). I haven’t found yet an ability to convert existing diagrams from SSMS to dbForge Studio. However I hope Devart developers will implement this feature in one of the following releases. All in all, editing Database Diagrams through dbForge Studio was a nice experience and allowed speeding-up the common database designing tasks. Download dbForge Studio now. Reference: Pinal Dave (http://blog.sqlauthority.com) Filed under: PostADay, SQL, SQL Authority, SQL Query, SQL Server, SQL Tips and Tricks, SQL Utility, T SQL

    Read the article

  • New Source Database Added for EBS 12 + 11gR2 Transportable Tablespaces

    - by John Abraham
    The Transportable Tablespaces (TTS) process was originally certified for the migration of E-Business Suite R12 databases going from a source database of 11gR1 or 11gR2 to a target of 11gR2. This requirement has now been expanded to include a source database of 10gR2 (10.2.0.5) - this will potentially save time for existing 10gR2 customers as they can remove on a crucial upgrade step prior to performing the platform migration. The migration process requires an updated Controlled patch delivered by the Oracle E-Business Suite Platform Engineering team, i.e. it requires a password obtainable from Oracle Support. We released the patch in this manner to gauge uptake, and help identify and monitor any customer issues due to the nature of this technology. This patch has been updated to now include supporting 10gR2 as a source database. Does it meet your requirements?Note that for migration across platforms of the same "endian" format, users are advised to use the Transportable Database (TDB) migration process instead for large databases. The "endian-ness" target platforms can be verified by querying the view V$DB_TRANSPORTABLE_PLATFORM using SQL*Plus (connected as sysdba) on the source platform:SQL>select platform_name from v$db_transportable_platform;If the intended target platform does not appear in the output, it means that it is of a different endian format from the source. Consequently. database migration will need to be performed via Transportable Tablespaces (for large databases) or export/import.The use of Transportable Tablespaces can greatly speed up the migration of the data portion of the database. However, it does not affect metadata, which must still be migrated using export/import. We recommend that users initially perform a test migration on their database, using export/import with the 'metrics=y' parameter. This will help identify the relative amounts of data and metadata, and provide a basis for assessing likely gains in timing. In general, the larger the amount of data (compared to metadata), the greater the reduction in downtime that can be expected from using TTS as a migration process. For smaller databases or for those that have relatively small data compared to metadata, TTS will not be as beneficial for cross endian migration and the use of export/import (datapump) for the whole database is recommended. Where can I find more information? Using Transportable Tablespaces to Migrate Oracle E-Business Suite Release 12 Using Oracle Database 11g Release 2 Enterprise Edition (My Oracle Support Document 1311487.1) Oracle Database Administrator's Guide 11g Release 2 (11.2) Related Articles Database Migration using 11gR2 Transportable Tablespaces Now Certified for EBS 12 New Source Databases Added for Transportable Tablespaces + EBS 11i 10gR2 Transportable Tablespaces Certified for EBS 11i Migrating E-Business Suite Release 11i Databases Between Platforms Migrating E-Business Suite Release 12 Databases Between Platforms

    Read the article

< Previous Page | 36 37 38 39 40 41 42 43 44 45 46 47  | Next Page >