Focused crawls are collections of frequently-updated webcrawl data from narrow (as opposed to broad or wide) web crawls, often focused on a single domain or subdomain.
This was our Final Project for Distributed Computing. In this we had to create a distributed system that will use the Brute Force Algorithm to crack a password from the etc/shadows file of Ubuntu as all the passwords are stored there. We used OpenMP and MPICH for this implementation.
In this task, I wrote the code using both OpenMP and MPI inorder to parallelize it. What it does is that the master process distributes a chunk of number equally among the slave processes. Each slave process then searches for the number to find. When one slave process finds the number, it sends a message to the master process who then sends a "terminating message" to the slave processes who then abort their search.
In this project, I experienced parallel programming with C++ using MPI library. I implemented a parallel algorithm for image denoising with the Ising model using Metropolis- Hastings algorithm.
Huh, while writing the spack package I wonder: will we make MPILander to be CUDA-aware? xD
Probably just depends on the host system and if the kernel supports unified memory.