I was reviewing some documents for making my database perform better and I came across with "OLAP" pre-aggregation term. I was wondering if OLAP is a tool or or methodology or approach. For example my DBMS is postgresql and I am working on a big databse. To speed up I have to use some aggregation and pre-aggregation methods. How OLAP can be helpful?
OLAP is a database role. When storing OLAP data in the db, typically you aren't running live transactional information off the db, but rather keeping it around for analytical and business intelligence reasons.
It isn't a tool. It isn't an approach either, since some approaches are needed for OLAP but some are helpful together in transactional environments as well.
In general you shouldn't think about speeding up an application by incorporating OLAP into it. Instead you would look at separating out reporting functions into a separate db server, and import the data periodically, and then separate data feeds from operation data stores, etc. This is a very different field than transactional application development.
Related
I'm getting familiar with the greenplum solution concepts, and trying to understand whether, and if so, when the organisation I work for should use this solution. Our conceptual idea is to setup a kind of central 'datastore' suitable for both OLTP and OLAP access.
My research: this article suggests Greenplum is more suitable for OLAP, and PostgreSQL for OLTP. But I also read about Greenplum improvements for OLTP processing. And in favour of Postgresql, there are also articles like this that suggest that OLAP (eg, a datawarehouse implementation) can be done by means of Postgresql.
So my question is: how to move forward, and what are the main criteria to decide? For example, in case we now have a just a few TB's (1-5), start with a Postgresql cluster (for OLTP+OLAP), and when data volumes grow, move on to Greenplum? Or start straight away with Greenplum?
maybe use postgres if it can handle your use case. If you have you have too much data and need to finish reports and analytics faster; change to greenplum
i have a problem where in i need to store the users's address data which can come from different vendors in different format. once i have the data i need to do some cleaning and wrinkling and run the de-duplication process to get the clean structured data. once the data is clean, i may have to pick the different attributes of address from different vendors based on some complex logic which is not defined yet. my question is
1) which database i should use i.e. NOSQL database family like document/keyvalue/dynamoDB etc or RDBMS with MPP database like redshift or azure data warehouse
2) NOSQL DB like mongoDB provide the flexibility of schema but at the same time the queries or de-duplication process is not something inbuilt for these databases.
if anyone can guide me on this i shell be very thankful for him
Thanks
Atul
There is a web application which is running for a years and during its life time the application has gathered a lot of user data. Data is stored in relational DB (postgres). Not all of this data is needed to run application (to do the business). However form time to time business people ask me to provide reports of this data data. And this causes some problems:
sometimes these SQL queries are long running
quires are executed against production DB (not cool)
not so easy to deliver reports on weekly or monthly base
some parts of data is stored in way which is not suitable for such
querying (queries are inefficient)
My idea (note that I am a developer not the data mining specialist) how to improve this whole process of delivering reports is:
create separate DB which regularly is update with production data
optimize how data is stored
create a dashboard to present reports
Question: But is there a better way? Is there another DB which better fits for such data analysis? Or should I look into modern data mining tools?
Thanks!
Do you really do data mining (as in: classification, clustering, anomaly detection), or is "data mining" for you any reporting on the data? In the latter case, all the "modern data mining tools" will disappoint you, because they serve a different purpose.
Have you used the indexing functionality of Postgres well? Your scenario sounds as if selection and aggregation are most of the work, and SQL databases are excellent for this - if well designed.
For example, materialized views and triggers can be used to process data into a scheme more usable for your reporting.
There are a thousand ways to approach this issue but I think that the path of least resistance for you would be postgres replication. Check out this Postgres replication tutorial for a quick, proof-of-concept. (There are many hits when you Google for postgres replication and that link is just one of them.) Here is a link documenting streaming replication from the PostgreSQL site's wiki.
I am suggesting this because it meets all of your criteria and also stays withing the bounds of the technology you're familiar with. The only learning curve would be the replication part.
Replication solves your issue because it would create a second database which would effectively become your "read-only" db which would be updated via the replication process. You would keep the schema the same but your indexing could be altered and reports/dashboards customized. This is the database you would query. Your main database would be your transactional database which serves the users and the replicated database would serve the stakeholders.
This is a wide topic, so please do your diligence and research it. But it's also something that can work for you and can be quickly turned around.
If you really want try Data Mining with PostgreSQL there are some tools which can be used.
The very simple way is KNIME. It is easy to install. It has full featured Data Mining tools. You can access your data directly from database, process and save it back to database.
Hardcore way is MADLib. It installs Data Mining functions in Python and C directly in Postgres so you can mine with SQL queries.
Both projects are stable enough to try it.
For reporting, we use non-transactional (read only) database. We don't care about normalization. If I were you, I would use another database for reporting. I will desing the tables following OLAP principals, (star schema, snow flake), and use an ETL tool to dump the data periodically (may be weekly) to the read only database to start creating reports.
Reports are used for decision support, so they don't have to be in realtime, and usually don't have to be current. In other words it is acceptable to create report up to last week or last month.
I have recently started getting familiarized with NoSQL (HBase). I am definitely a noob.
I was investigating about ORMs and high level clients which can be used on HBase and came across a few.
Some ORM libraries like Kundera are providing SQL like data query functionality. I am finding this a little counter intuitive.
Can any one help me understand why we would again need SQL like querying if the whole objective was to move away from it?
Also can anyone comment on your experiences with ORMs for HBase? I looked at a few of them from http://wiki.apache.org/hadoop/SupportingProjects and started looking at Kundera.
Another related question - Does data query with Kundera run map reduce jobs internally?
kundera or Spring data might provide user friendly ORM layer over NoSQL databases, but the underlying entity model still has to be NoSQL friendly. This means that NoSQL users should not blindly follow RDBMS modeling strategies but design ORM entities in such a way so that all NoSQL capabilities can be used.
As a thumb rule, the kundera ORM entities should be designed using query-first strategy where first the queries need to defined so as to create primary keys and also ensuring that relationship model is used as minimal as possible. Querying on random columns and full scans should be avoided and so data might have to be replicated across entities for reducing multiple entity look ups. Also, transactions management needs to be planned. FYI, kundera does not support transactions(beyond single row TX supported by Hbase/Cassandra).
Reason for using Kundera:
1) If looking for SQL like support over HBase. As it is build on top of HBase native API, so it simply transforms these SQL queries in to corresponding GET or PUT method calls.
2) Currently it support HBase-0.20.6 only. Kundera-2.0.6 will enable support for HBase 0-90.x versions.
3) Kundera does not do sometihng out of the box to provide map reduce over SQL like queries. However support for such thing will be provided in Kundera-2.0.6 by enabling support for Hive native queries only!
It is totally JPA compliant, so no need to learn something new. It simply hides complexity at developer level with very minimal effort.
SQL like querying is for developement ease, quick developement, less error prone and reusability ofcourse!
-Vivek
Does anyone have experience of using PostgreSQL for an OLAP setup, using cubes against the database etc. Having come across a number of idiosyncracies when using MySQL for OLAP, are there reasons in favour of using PostgreSQL instead (assuming that I want to go the open source route)?
There are a number of data warehousing software vendors that are based on Postgresql (and contribute OLAP related changes back to core fairly regularly). Check out https://greenplum.org/. You'll find that PG works a lot better (for nearly any workload, OLAP especially) than MySQL. Greenplum and other similar solutions should work a bit better than PG depending on your data sets and use cases.
PGSQL is much better suited for Data Warehousing compared to MySQL. We had thought initially to go with MySQL, but it performs poorly in aggregations if data grows to a few million rows. PGSQL performs almost 20 times faster in caparison with MySQL for 20 million records for a single fact table on same hardware setup. If for some reason you choose to go with MySQL, then you should use MyISAM storage engine for fact tables rather then InnoDB; you will see slightly better performance.