FOSSPicks
Storage monitor
Parallel Disk Usage
There are lots of tools for looking at disk usage, and we've covered quite a few of them over time in these pages. On the command line, many of us simply resort to piping the output from du
into a variety of other tools or use either dutree
or ncdu
for a more visual approach. The only problem is that all of these tools can take a considerable amount of time to grind through your data before they can produce their output. And you invariably want to change the search after getting the results, which means going through the same process again. Parallel Disk Usage (pdu) has been developed to solve this problem.
Parallel Disk Usage is orders of magnitude faster than any of the alternatives. On modern systems with SSD storage and multiple cores to spare, we barely noticed the difference in output time between the humble ls
and pdu
. This is despite pdu displaying all the files and directories beneath your chosen destination, complete with lines to show their relationships, their size, and an incredibly useful bar chart that gives you a quick overview of which files and folders are taking the most space. The chart defaults to showing a percentage value for how much of the destination space a specific file or directory is taking. It's a brilliant way to find unexpected resource sinks, such as hidden cache directories or forgotten virtual machines. There are configuration options to change its width, measure blocks rather than bytes, limit trawling depth, and even output the results as JSON. The "parallel" in its name refers to the mechanism that makes pdu so quick – or "blazingly fast," as the project puts it. This mechanism harnesses the parallelized nature of the Rust programming language to make best use of the multiple cores in your CPU, which is why pdu results are delivered with the same speed and agility as ls
.
Project Website
https://crates.io/crates/parallel-disk-usage
File sharing
Buy this article as PDF
(incl. VAT)