I am designing a multi-tenant database. Many actions that are needed on a per-tenant basis have been written as stored procedures. Almost all of them use dynamic SQL execution since I have not found a way to differentiate between the schemas other than with dynamic statements.
EXEC( 'SELECT * FROM ' + #SchemaName + '.Contacts' )
Is there any way to create a variable representing the schema so I can call the select statements without building them dynamically?
SELECT * FROM #TheSchema.Contacts
SQL Server 2008
FWIW I have found it much more useful to separate customers by database than by schema. Reasons?
All schemas can have identical procedures, thus derive them from model instead of having to create anything. Deployment to multiple databases is no more complex than deployment to multiple schemas, actually less so I'd argue, and the only difference is the application would have to know which database to reference instead of which schema.
Different customers can have different recovery models, be backed up on different schedules, etc. This can improve your backup/restore plan especially as the data set gets larger.
Having each customer in a separate database makes it very easy to move a tenant to a different server if they get too big for the current one. Extracting the data and objects for their schema only will be quite convoluted.
Some customers may legally and/or contractually require that you not store their data in the same database as other customers.
A simpler solution is to keep the stored procedure in each of the tenant databases and execute it inside them.
This question is a bit old, but I am looking into this option as a cost savings measure - since many cloud providers charge per database and the costs associated with storage are less - especially useful for multi-tenant small databases.
From what I understand, a default schema can be set for the login, thus, setting a schema per login would automatically use that schema for all queries and sprocs. Something to consider for those that comes across this question.
Related
SQL Server and Oracle terminology -
In SQL Server If I have two applications and want to keep the database completely separate, I could simply create 1 database for each application therefore I end up with 2 databases.
If I wanted to do the same thing in oracle, what do I need to create?
- create a new "Databases"? "Instance", "Schema", or "Tablespace" per application?
(Note, these two applications is the same application used by two different companies, that do not share data!)
Reference: http://www.codeproject.com/Tips/492342/Concept-mapping-between-SQL-Server-and-Oracle
Having worked with SQL Server a lot in the past, I have sympathy with trying to figure out how Oracle organizes things as I struggled with the same thing. My comments below are from SQL Server 2000 and 2003 so forgive me if things have changed since then.
Previous responders have been helpful. I think one problematic assumption here is that there is an exact "level" equivalency between SQL Server and Oracle. What I mean by "level" is something that occupies the same space in the hierarchies that you have diagrammed above (and which, btw, I think are a good place to start but might need a bit of editing in a couple of places, for example how you have diagrammed "user" and "schema" in the Oracle hierarchy, I might put them side-by-side.) I do not think these concept "levels" match exactly between the DB platforms.
A schema in Oracle is somewhat equivalent to a separate database in SQL Server but not entirely.
I would say that the "walls" -- not an exact technical term but oh well -- between databases in SQL server are a bit higher than the "walls" between schemas in Oracle. Others might disagree but here is my reasoning:
a. A schema in Oracle is a purely logical construct. It denotes who has ownership of objects. It has nothing to do with the physical location or layout of the objects. A tablespace (orthagonal concept, as noted by a previous poster) indicates the physical location of objects. A tablespace can hold objects that are in multiple schemas and vice versa. In SQL Server these two concepts are sort of merged into one -- a database is both tablespace and schema, more or less, although in some respects within a DB in SQL Server you then have multiple owners with various object ownership. This can get a bit confusing because as I remember (it's been a couple of years) if not using NT Authentication the users are defined at the server level and then have to "link" to the users in the individual DBs.
b. I remember finding it easier, or at least a bit simpler, to assure myself that users to two separate DBs in SQL Server had no access to the relative other user's DB than I have found it in Oracle.
c. Because a DB in SQL server represents both physical storage and logical ownership, you can detach the DB and move it to another SQL Server Instance and attach it. You can't do this with a schema in Oracle. I mean, you can datapump the data out or back it up or whatever to another server and another schema, but that all takes at least some scripting and such or at least a fair amount of clicking in Enterprise Manager. It doesn't give you the one-click "Detach DB" option that you have in SQL Server which makes it a lot easier to get the idea that SQL Server DBs are units that you can more-or-less move back and forth between databases.
To sum things up, I think either option would work. That is, 1) Create two separate instances of Oracle with one schema in each instance for each application, or 2) Create two separate schemas in one Oracle instance.
There are pros and cons for each option. Option 1 is probably going to be more work to set up and configure but will also give you more separation, independence, ability to have separate hardware, etc., for each DB. Option 2 will be quite a bit simpler but gives you less separation between the data and greater risk of configuration screw-ups or other things allowing users of one schema to access the other. It also means you have to be a bit more careful that someone writing a query accessing data in one schema doesn't use all the CPU and IO resources and starve a user on the other schema.
Also, yes, you could use pluggable databases in 12c. However, given the fact that you need to ask these questions (no shame, just pointing out where you're at) makes me hesitant about recommending what can easily be a more complex setup.
TL;DR -- SQL Server isn't Oracle and Oracle isn't SQL Server. Either option works and there are pros and cons to each.
If you're using 12.1 or later with the multitenant option, you could create separate pluggable databases in a single container database. The other option, which works in any version of Oracle, would be to create a separate schema. It would be possible, as well, to create a separate database, though that is generally not the preferred approach unless you have a particular need to do things like upgrade the database that one application is using without affecting the other.
Creating a Database
If you create a separate database, you'd end up with complete separate memory structures (i.e. the SGA and PGA for each database would be separate) as well as a completely separate set of background processes (each database would have its own log writer process(es) for example). That is a very heavyweight option-- you can't have too many databases on a single server before you start having a lot of contention for RAM, for scheduling all the background processes, etc. It does provide for the maximum separation between different applications-- each database can be running a different version of Oracle with a different set of initialization parameters-- but this also tends to increase the complexity of managing the environment. This generally only makes sense when you have third party applications that require a specific version of the database or a specific set of initialization parameters.
Creating a Schema
If you create a separate schema, you still have a single database so the two schemas are sharing the same memory structures (competing with each other for space in the SGA's buffer cache, for example), initialization parameters, etc. You have to exercise a modicum of planning to ensure that that the two don't interfere with each other-- you'd probably want to make sure that nether application creates public synonyms or at least that they won't wan to create the same public synonym as the other application-- but this is generally pretty trivial.
Creating a Pluggable Database
This only works in 12.1 and only if you have the multitenant option. This is the most similar to the SQL Server concept of creating a new database for each application.
You should create a new instance (schema) on the same database, where the schema in oracle is the same as the SQL server database
I have a scenario, my application is a SAAS based app catering to multiple clients. Data Integrity to clients is very essential.
Is it better to keep my Tables
Client specific
OR
Relational Tables
For Ex: I have a mapping table with fields MapField1,MapField2. I need this kind of data for each client.
Should I have tables like MappingData_
or a Single Table with mapping to the ClientId
MappingData with Fields MapField1,MapField2,ClientId
I would have a separate database for each customer. (Multiple databases in a single SQL Server instance.)
This would allow you to design it once, with a single schema.
No dynamically named tables compromising test & development
Upgrades and maintenance can be designed and tested in one DB, then rolled out to all
A single customer's data can be backed-up, restored or dropped exceedingly simply
Bugs discovered/exploited in one DB won't comprise the integrity of other DBs
Data access (read and write) can be managed using SQL Logins (No re-inventing the wheel)
If there is a need for globally shared data, that would go in another database, with it's own set of permissions for the different SQL Logins.
The use of a single database, with all users in it is my next best choice. You still have a single schema. But you don't get to partition the customers' data, you need to manage access rights and permissions yourself, and a whole host of other additional design and testing work.
I would never go near dynamically creating new tables for additional customers. A new table name means all your queries need to be updated with the new table name, and a whole host of other maintenance head-aches.
I'm pretty much of the opinion that if you want to create tables dynamically during the Business As Usual use of an application/service, you've designed it badly.
SO has a tag for the thing you're describing: "multi-tenant".
Visualize the architecture for supporting a multi-tenant database application as a spectrum. At one extreme of the spectrum is "shared nothing", which means each tenant has its own database. At the other extreme of the spectrum is "shared everything", which means tenants share tables, and each row in each table belongs to one tenant. (Each row contains a tenant identifier.)
Terminology seems to overlap, so read carefully. What one writer means by shared schema might be identical to what another writer means by shared everything.
This SO answer, also written by me, describes the differences and the tradeoffs in terms of cost, data isolation and protection, maintenance, and disaster recovery. It also links to a fairly good introductory article.
I've currently got a "Web" edition SQL Azure server with on database on it. I want to put another database on there but am unsure how the costing works. Will I need to pay the £9.99 per database or database server?
Does it make more sense just to set up a couple of different schemas in my existing database to try and reduce costs?
You pay per database. Creating schemas may make sense if cost is your concern. I have seen this done multiple times. However keep in mind that a few SQL Server/Azure features are schema independent. For example user-defined statistics and roles are schema independent. So as long as you don't use these features you should be good with a schema-based separation.
I have a database server with few main databases, and few dozens of small ones.
These small databases are kind of intermediary/staging databases for data import from various sources into main database. Data import is a daily task. They are all quite similar in structure as the implementation of these data imports are similar, so basically they have a configuration tables, which define mapping, conversions etc, and the data tables, which contain the results of the import.
Some time ago there have been only the handful of small ones, but now I have more then 20 of them will grow further with the number of supported data feeds.
I have just migrated all the server environment to SQL Server 2008, and having some time now for clean-up/refactoring, I am thinking to merge all of data-import databases into just one database, and use database schema to separate them.
Question-0: Any other ideas for the described situation?
Question-1: Shall I change from a separate database to a separate schema?
Question-2: !!!: Any tricky thing to be careful about in database schema implementation?
Edit-1: highlighted question-2 as the most 'unanswered' currently.
In your instance, I would probably put merge the databases into one. I don't really see a reason to have them separated, and merging them will reduce the amount of work you have to do to support backups etc. If you were importing data from a data source once and then never using the staging tables again, I could see the reason to bring up separate databases to handle the data transformation. Since you use these tables on an ongoing basis, I would much rather keep them together so that I only have to go to one place to find the full end to end state of the production data and the data load states.
2008 is really good at handling database partitioning too, if the db gets too large, or you need to separate data for security reasons you get the benefit of having a single db with the advantages like having several smaller ones. You won't get that with multiple smaller dbs.
When we migrated we had a very similar situation and I ended up moving everything into one some-what large Importing database like you have hinted towards. We did not, however, separate them using schemas.
Because the database is the unit of referential integrity and backup, if you are bringing in large amounts of data for staging which does not need to be backed up on the same schedule, it might be easiest to keep it in a separate DB.
You can use a single DB with multiple file groups and different backups, but it will require a lot more design.
The basic factors this will depend on are: recovery model, backup objectives, usage patterns and amount of effort to design and maintain your file group design.
All the prior answers work for me, particularly your comment about selectively combining databases -- if some are very busy, very large, or process sensitive data, you might want to keep them separate, or in separate groupings. This would make it easier to configure backups/restores and disk/drive allocation (give the busy ones their own set of spindles).
Like possibly most database developers, I have dealt almost exclusively with objects in the dbo schema, but I have done some recent work with other schemas. The main gotcha I've encountered is remembering to always specify the schema when referring to any database object. Never assume that any given connection will reference an object in the schema you want it to--always be clear and precise!
I would put all your import staging tables in one database separate from your regular production databse as the backup needs may be very different. This database should also contains things like your configuration management for SSIS packages, any logging tables, any import metadata tables (we keep track of every run of the imports and the status of that run as well as a bazillion other things about the import like the filename, the normal file size, etc. Comes in handy for researching problems and for adding checks to the processing. We usea a schema that is by client and then an additional schema for objects realted to the importing/exporting process (logs, meta data etc.)
In my project I have two separate DBs on the same server. The DBs are self-sufficient except for three columns in DB "B" that need to be accessed in DB "A".
Are there performance considerations if I were to have a stored proc in A that accessed three columns from B directly?
Currently, we run a nightly job to import the needed data from table B to table A, so that the stored proc isn't going out of the scope of A.
Is that the best method?
Are cross DB stored procs within best practices?
There should be no problem since the databases are on the same server. Usually problems occur when you do this with linked servers and you could run into network latency
To clarify the other posters comments.
There is no "direct" negative performance impact when using cross database access via stored procedures. Performance will be determined by the underlying architecture of the individual databases, i.e. indexes available, physical storage locations etc.
This is actually quite a common practice and so long as you follow standard query tuning principals you will be just fine.
Yes, what you're currently doing - i.e. replication - is the "correct" thing to do.
When referencing data in another database, you can't use referential integrity, data constraints and lots of the other good stuff that make an RDBMS a good tool to use.
Accessing the other database direcly ties the databases together - they MUST exist on the same server, for all time. You may run into badwidth issues using linked servers and executing the proc on demand.
Replication gives you far more flexibility.
It all depends on the data you are referencing and whether or not indexes are set up to support this direct access. The best thing I can tell you is to create the query, run it, and see if the performance is good enough.
If the performance is not satisfactory then run the query again and have your management tool generate a execution plan so you can identify the bottleneck.
We do this all the time. As long as they are on the same server there is no problem. If you have a requirement that the data must be in table a on database a before it can go into table B on database b, you will need to write a trigger to check as foreign key relationships can't be set across databases.