Denormalisation -having your cake and eating it

2010

TL; DR

Why and when denormalisation makes sense.

Session Details

Used properly, normalisation brings huge advantages. It optimises storage (each piece of data is stored only once), it removes an entire class of update, insert and delete anomalies and it improves data integrity. What more could we ask for? Well, performance can be an issue. Normalised databases often have a reputation for poor performance. This talk will examine the role of normalisation on performance and focus on effective ways we can denormalise data and yet retain the data integrity that normalisation brings. This talk compliments that by Tony Rogerson beautifully and will be given by Mark and Yasmeen Ahmed who works with him at the University of Dundee.

3 things you'll get out of this session

Mark Whitehorn's previous sessions

Graph databases - What, how and why

Graph databases – Origins, how they work, strengths and weaknesses

Inside the classic machine learning algorithms

Algorithms like Neural Nets, Deep Learning, SVM and KNN are dominating the Machine Learning world. You don’t need to know the maths behind them but you do need to know how they work in order to use them effectively. This talk will explain just that.

Evaluating Models: Get Ready to ROC

Introducing the ROC curve and its use in Azure ML.

Analytics Master Class-exploiting exotic patterns in data

There are some little-known but very useful ways of extracting information from data. This session will cover: • Monte Carlo simulations • Nyquist’s Theorem • Simpson’s paradox • Benford’s Law These will rock your world (they certainly rocked mine).

MDX and DAX-compare and contrast

This talk looks at MDX and DAX, examining their similarities and differences. It won’t turn you into an expert in either but it will help you to decide, given your particular career plans, if either or both are worth learning.