I have an OLTP database, and am currently creating a data warehouse. There is a dimension table in the DW (DimStudents) that contains student data such as address details, email, notification settings.
In the OLTP database, this data is spread across several tables (as it is a standard OLTP database in 3rd normal form).
There are currently 10,390 records but this figure is expected to grow.
I want to use Type 2 ETL whereby if a record has changed in the OLTP database, a new record is added to the DW.
What is the best way to scan through 10,000 records in the DW and then compare the results with the results in several tables contained in the OLTP?
I'm thinking of creating a "snapshot" using a temporary table of the OLTP data and then comparing the results row by row with the data in the Dimension table in the DW.
I'm using SQL Server 2005. This doesn't seem like the most efficient way. Are there alternatives?
Introduce LastUpdated into source system (OLTP) tables. This way you have less to extract using:
WHERE LastUpdated >= some_time_here
You seem to be using SQL server, so you may also try rowversion type (8 byte db-scope-unique counter)
When importing your data into the DW, use ETL tool (SSIS, Pentaho, Talend). They all have a componenet (block, transformation) to handle SCD2 (slowly changing dimension type 2). For SSIS example see here. The transformation does exactly what you are trying to do -- all that you have to do is specify which columns to monitor and what to do when it detects the change.
It sounds like you are approaching this sort of backwards. The typical way for performing ETL (Extract, Test, Load) is:
"Extract" data from your OLTP database
Compare ("Test") your extracted data against the dimensional data to determine if there are changes or whatever other validation needs to be performed
Insert the data ("Load") in to your dimension table.
Effectively, in step #1, you'll create a physical record via a query against the multiple tables in your OLTP database, then compare that resulting record against your dimensional data to determine if a modification was made. This is the standard way of doing things. In addition, 10000 rows is pretty insignificant as far as volume goes. Any RDBMS and ETL process should be able to process through that in a matter of no more than few seconds at most. I know SQL Server has DTS, although I'm not sure if the name has changed in more recent versions. That is the perfect tool for doing something like this.
Does you OLTP database have an audit trail?
If so, then you can query the audit trail for just the records that have been touched since the last ETL.
Related
I'm investigating data warehouses. And I have an issue about star schemas.
It's in
Oracle® OLAP Application Developer's Guide
10g Release 1 (10.1)
3.2.1 Dimension Table: TIME_DIM
https://docs.oracle.com/cd/B13789_01/olap.101/b10333/global.htm#CHDCGABE
To represent the hierarchy MONTH -> QUARTER -> YEAR, we need some keys such as: YEAR_ID, QUARTER_ID. But there are some things that I do not understand:
1) Why do we need field YEAR_DSC & QUARTER_DSC? I think that we can look up these values from YEAR & QUARTER TABLE. And it breaks 2NF.
2) What is the normal form that a schema in data warehouse needs to satisfy? (1NF, 2NF, 3NF, or any.)
NFs (normal forms) don't matter for data warehouse base tables.
We normalize to reduce certain kinds of redundancy so that when we update a database we don't have to say the same thing in multiple places and so that we can't accidentally erroneously not say the same thing where it would need to be said in multiple places. That is not a problem in query results because we are not updating them. The same is true for a data warehouse's base tables. (Which are also just queries on its original database's base tables.)
Data warehouses are usually optimized for reading speed, and that usually means some denormalization compared to the original database to avoid recomputation at the expense of space. (Notice though that sometimes rereading something bigger can be slower than reading smaller parts and recomputing the big thing.) We probably don't want to drop normalized tables when moving to a data warehouse, because they answer simple queries and we don't want to slow down by recomputing them. Other than those tradeoffs, there's no reason not to denormalize. Some particular warehouse design methods might have their own rules about what parts should be denormalized what amounts.
(Whatever our original database design NF is chosen to be, we should always first normalize to 5NF then consciously denormalize. We don't need to normalize or know constraints to update or query a database.)
Read some textbook basics on why we normalize & why we use data warehouses.
As far as I know, normalization is done to avoid inconsistency in the database.
By normalizing we:
reduce data redundancy, and
protect data integrity.
That's why most OLTP databases are in 3NF.
Different databases from OLTP come together in a data warehouse. (DWH, OLAP). DWHs are denormalized (1FN), and is obvious it has to be like that, because the main table of a DWH has hundreds of columns.
From that DWH we can build several data marts that we would later use for doing analysis with a BI reporting tool (Cognos, QlikView, BO .. )
The problem is that the data model for the BI report is not normalized.
Couldn't that be a problem for redundancy and data integrity for the report?
In OLAP systems (such as data warehouses), the key efficiency needs are in querying and data retrieval.
Therefore some of the design considerations are done in order to retrieve the information faster, even if the updates may be longer.
An example for such a model is a Star-Schema on which we denormalize data in a such way that all the data will be stored in a 1-join-hop distance.
Key elements such as transactions are located at the big table (Facts), with foreign keys pointing at the dimensions.
The dimension themselves are smaller, and may contain not-normalized data. For example an address dimension may store street, neighborhood and city data without normalizing it to 3NF.
There are for sure redundancy issues (you don't really have to store Day_of_Week per each date row) but it is insignificant (since storage is not a bottleneck in this scenario).
As per integrity - you face it only on updates (F.E. a less-realistic scenario of country change per State_Province in Dim_Store) , and in DWH update is a rare-case, where we allow ourselves to be inefficient.
Moreover - integrity is not enforced by the DB (or normalization) but by design and implementation of the ETL process.
Read more on Data Warehouses modeling
Regarding redundancy: some data warehouse engines like Amazon Redshift allow data compression that is very handy for denormalization. Let's say you have a table of sales events with 100M records and every sale has a city. In OLTP data model, you would have sales and cities with city_id connecting them. In OLAP data model with compression allowed it's much easier to have sales table with a text city attribute compressed. You'll be able to calculate sales by city without joining tables, and your city values won't occupy much disk space because they will be encoded.
More info about compression is in Amazon docs: Choosing a Column Compression Type
Regarding data integrity: you have to design your ETL routines to minimize the possibility of duplicate data and also run scheduled checks for duplicates based on criteria like this:
select count(*) from table;
select count(*) from (select distinct <expression> from table);
where is a list of columns which combination should be unique (your artificial primary key).
First things first, I am an amateur, self-taught ruby programmer who came of age as a novice engineer in the age of super-fast computers where program efficiency was not an issue in the early stages of my primary GIS software development project. This technical debt is starting to tax my project and I want to speed up access to this lumbering GIS database.
Its a postgresql database with a postgis extension, controlled inside of rails, which immediately creates efficiency issues via the object-ification of database columns when accessing and/or manipulating database records with one or many columns containing text or spatial data easily in excess of 1 megabyte per column.
Its extremely slow now, and it didn't used to be like this.
One strategy: I'm considering building child tables of my large spatial data tables (state, county, census tract, etc) so that when I access the tables I don't have to load the massive spatial columns every time I access the objects. But then doing spatial queries might be difficult on a parent table's children. Not sure exactly how I would do that but I think its possible.
Maybe I have too many indexes. I have a lot of spatial indexes. Do additional spatial indexes from tables I'm not currently using slow down my queries? How about having too many for one table?
These tables have a massive amount of columns. Maybe I should remove some columns, or create parent tables for the columns with massive serialized hashes?
There are A LOT of tables I don't use anymore. Is there a reason other than tidiness to remove these unused tables? Are they slowing down my queries? Simply doing a #count method on some of these tables takes TIME.
PS:
- Looking back at this 8 hours later, I think what I'm equally trying to understand is how many of the above techniques are completely USELESS when it comes to optimizing (rails) database performance?
You don't have to read all of the columns of the table. Just read the ones you need.
You can:
MyObject.select(:id, :col1, :col2).where(...)
... and the omitted columns are not read.
If you try to use a method that needs one of the columns you've omitted then you'll get an ActiveModel::MissingAttributeError (Rails 4), but you presumably know when you're going to need them or not.
The inclusion of large data sets in the table is going to be a noticeable problem from the database side if you have full table scans, and then you might consider moving these data to other tables.
If you only use Rails to read and write the large data columns, and don't use PostgreSQL functions on them, you might be able to compress the data on write and decompress on read. Override the getter and setter methods by using write_attribute and read_attribute, compressing and decompressing (respectively of course) the data.
Indexing. If you are using postgres to store such large chucks of data in single fields consider storing it as Array, JSON or Hstore fields. If you index it using the gin index types so you can search effectively within a given field.
There is a Java Swing application which uses an Informix database. I have user rights granted for the Swing application (i.e. no source code), and read only access to a mirror of the database.
Sometimes I need to find a database column, which is backing a GUI element (TextBox, TableField, Label...). What would be best approach to find out which database column and table is holding the data shown e.g. in a TextBox?
My general approach is to capture the state of the database. Commit a change using the GUI and then capture the state of the database again. Then I need to examine the difference. I've already tried:
Use the nrows field of systables: Didn't work, because the number in nrows does not seem to be a realtime representation of the row count.
Create a script with SELECT COUNT(*) ... for all tables: didn't work because too many tables (> 5000). Also tried to optimize by removing empty tables, but there are still too many left.
Is there a simple solution that I'm missing?
Please look at the Change Data Capture API and check if this suits your needs
There probably isn't a simple solution.
You probably need to build yourself a map of the database, or a data dictionary for it. It sounds as though you can eliminate many of the tables from consideration since they're empty — at least for a preliminary pass. If you're dealing with information in a text box, the chances are it is some sort of character data; you can analyze which (non-empty) tables which contain longer character strings, and they'd be the primary targets of your searches. If the schema is badly designed with lots of VARCHAR(255) columns even though the columns normally only hold short strings, life is more difficult. Over time, you can begin to classify tables and columns so that you end up knowing where to look for parts of the application.
One problem to beware of: the tabid in informix.systables isn't necessarily as stable as you'd like. Your data dictionary needs to record its own dd_tabid for the table it describes, and can store the last known tabid from informix.systables, but it needs to be ready to find a new tabid value on occasion. You should probably only mark data in your dictionary for logical deletion.
To some extent, this assumes you can create a database in which to record this information. If you can't create an Informix database, you may have to use something else (MySQL, or SQLite, perhaps) to store the data dictionary. Alternatively, go to your DBA team and ask them for the information. Unless you're trying something self-evidently untoward, they're likely to help (but politics can get in the way — I've no idea how collegial your teams are).
I am using json_extract_path_text function to extract values from JSON. As row data increases, query takes long time to run and fails for some time.
Is there a way to reduce query execution time or improve josn_extract_path_text function
The solution is: store your data in tabular format instead of JSON. JSON is not a good choice for storing larger data sets because, by storing disparate data in a single column, JSON does not leverage Amazon Redshift’s column store architecture. Or alternatively change you node type to bigger one.
Redshift being a columnar store, storing data in JSON format would not speed up queries on it. This would work on a document model NOSQL database, but not on RedShift. To make RedShift queries efficient, distribution style(even for the scenario where data does not follow a speicifc order or is random) of the tables are important, based on the number of clusters you have. Also, Distribution key on the primary key column(in an otherwise RDBMS model), and Sort Key on the same would help you in Joins(it would use the Sort Merge join instead of the longer Hash Join).
For more details about this, do have a look at the documentation. RTFM is your friend here.