Evaluation of Caching Frameworks

By Rob Tan

To build a system for responsive performance, one often employs the technique of caching data in a fast in-memory store as opposed to hitting the database. At Nomis Solutions, we wanted to compare the performance of various frameworks, considering both local and distributed caches and including internal in-memory caching using a HashMap. We also wanted to compare the performance of no caching at all (i.e. hitting our MongoDB database directly).

Read More »

MongoDB JAVA Driver: QueryBuilder Class

By Deep Pancholi

I was recently assigned a task where I had to convert a SQL like where clause query into Mongo query on the fly. There are some good drivers available to do the same but they are not free. The best example is UnityJDBC but it was a bit expensive for our use case. The query needed to be super simple and we made the following assumptions:

Read More »

Anscombe’s Quartet and the Case for Visualizations

By Niranjan Shetty

In his classic paper on Graphs in Statistical Analysis published in 1973, F. J. Anscombe presents what we now know as the Anscombe’s quartet – a set of 4 datasets with identical statistical properties that are very different when plotted. Anscombe’s quartet highlights the fallacy of relying purely on statistics, and emphasizes the importance of visualization. Anscombe’s paper begins with one of the shortest abstracts I have encountered in a technical paper.

Read More »

Simpson's Paradox

By Niranjan Shetty

Simpson's paradox (no relation to the animated sitcom) is the case where a reversal in the direction of an association is observed when an additional explanatory variable is taken into account. A popular example is the Berkeley Gender Bias case where the admission figures for 1973 showed that men were more likely to be admitted (44% of all male applicants) relative to women (35% of all female applicants) - a statistic based on which the University of California, Berkeley was sued. However, when the admission rates were broken down by departments, it was observed that there was a small but statistically significant bias in favor of women. Women tended to apply to more competitive departments (with low admission rates), while men tended to apply to less competitive departments (with higher admission rates). As a result, the aggregated results, men had a higher admission percentage than women. The paradox demonstrates some of the challenges in relying too much on patterns derived from aggregate data viewed along a single dimension.

Read More »
COMMENTS