Data warehouse security is vital for organizations that take the huge step of collating all their important data in one place. Unauthorized access to data warehouses can have devastating consequences, ranging from compromised customer information to exposing of high-level business intelligence that gives an enterprise its competitive advantage.

This article introduces warehouse security, discusses some of the challenges involved in securing a data warehouse, and offers some best practices for data warehouse security, including the use of FIPS 140-2 certified software and hardware..

This article describes how CoolaData implemented Druid to provide a real time infrastructure for its behavioral analytics platform, and discusses some benefits and limitations of Druid..

Written by Yair Harel, Senior Software Architect at CoolaData

In 2007 when we hosted the first The Finance and Investment Forum event in Cass Business School in London the data centre and cloud industry was still, in a sense, in it’s infancy. Even then few predicted not only the sector expansion but the technological advances and energy innovation that would come to recent prominence..

Imagine you need a report with data from an OLAP sales cube, and have to include data from a table of the relational data source, or from an external database.

With a linked server you can construct a MDX query in the SQLServer instance where you have the cube, joining with an external database to complete the data with information from tables of the relational database..

With SQL Server, when we forget or we lose the password of the DBA user 'sa', and we are the administrators of the database, we have a little problem.

It's easy to forget this password because probably we have our own user administrator and don't use to log in with the sa account. We also could have inherited the administration of a SQL Server database, with any documentation about the sa account or we coul'd have installed a database and deleted the login with admin privileges without knowing the password of the sa user.

For all this situations there is a better solution than reinstalling the database..

If you have to change the type of a key column on SQL Server, and the new column type has to be integer, and even identity, you can do it by using some criteria in order to get the rows ordered (PK= index clustered= order by in physical disk by this column)..

We will change the type of the PK column of a table in a sample with 2 typical invoice tables master-detail where the key columns are nchar. 'invheader.invoiceid' is the primary key of the master table and 'invlines.lineid' is the primary key of the detail table..

We could need to access to MySQL from SQL Server.

An easy way to do it is creating a linked server that uses an ODBC connection..

With almost 350 pages, 'Learning Hadoop 2', I was very pleasantly surprised!

This is a book that chapter by chapter, provides a guide with the appropriate depth in the apprenticeship of this technology, which is today one of the most flourishing. It has a lot of examples that we can execute and analyze..

The outliers, or extreme values, are a usual topic when analyzing data, regardless its data source. Here I present a didactic and visual approach on treating with them. R code is available to download.

Living longer, living better? 

Following analysis is made with R and the Google Vis library for plotting the interactive maps. 
It's equally important to measure the longer living as well as its quality..