Quirks with parfor vs. for
Parallelizing loops with Matlab's parfor might generate unexpected results. Users beware!
Performance
Afterthoughts on implicit expansion
The new implicit expansion feature of Matlab R2016b can break user code in unexpected ways.
Speeding up Matlab-JDBC SQL queries
Fetching SQL ResultSet data from JDBC into Matlab can be made significantly faster.
Zero-testing performance
Subtle changes in the way that we test for zero/non-zero entries in Matlab can have a significant performance impact.
A few parfor tips
The parfor (parallel for) loops can be made faster using a few simple tips.
rmfield performance
The performance of the builtin rmfield function (as with many other builtin functions) can be improved by simple profiling.
Viewing saved profiling results
We can easily use saved profiling results to analyze, view and compare profiling results of multiple runs.
Convolution performance
Matlab's internal implementation of convolution can often be sped up significantly using the Convolution Theorem.
Callback functions performance
Using anonymous functions in Matlab callbacks can be very painful for performance. Today's article explains how this can be avoided.
Solving an mput (FTP) hang problem
Matlab may hang when using passive FTP commands such as mput and dir. A simple workaround is available to fix this.