This document summarizes work on estimating sparse inverse covariance matrices using graphical lasso. It discusses how graphical lasso uses an L1 regularization and coordinate descent algorithms to efficiently estimate sparse inverse covariance matrices. The new GLASSO R package was developed that is 30-4000x faster than existing methods for estimating sparse graphical models on large datasets with thousands of nodes and parameters. Future work aims to apply this approach to even larger datasets where the number of parameters exceeds the number of samples.