2 ms·
I would be interested in hearing more about how you log / see this - a real world case always beats benchmarks
by lifeisstillgood 6y ago
I would be interested in hearing more about how you log / see this - a real world case always beats benchmarks
- bayindirh 6y agoActually, my method is very simple. Every 10 minutes I log the Total LBAs Written attribute from the SSD to a log file alongside a long time stamp. Periodically I graph the data with GNUPlot. There's a slight slope most of the time. In some regions there are jumps. These are generally when I seriously work with my software, generating logs and other output. System is on another SSD and swap is on a high performance HDD so, they do not affect the graphs I'm getting. That jumps helped me to catch two KDE bugs. One was an isolated case with akonadi. Other one was reported and possibly KRunner's bookmarks extension will see some more revisions to eliminate disk trashing. Addendum: When I have spare time, I'd write a small HTML/JS file to get the data from log file and graph it interactively. A both fun and useful project.