What is the use of sysctl.d/99-powersave.conf for an SSD drive?

[hjdskes]

What is the use of sysctl.d/99-powersave.conf for an SSD drive?

[Hjdskes]

That depends on what you mean with "not necessary", I suppose. If your goal is power saving, then you are probably correct in that it does not save as much power on an SSD as it does on an HDD (haven't tested how much of a difference, if any, it makes. I assume it would be negligible). However, less writing improves the lifespan of your SSD so in that sense you could deem it necessary (although, with modern SSDs, this is also not much of an issue anymore).

In any case, it won't hurt to have them listed. Most of the entries in that file are currently commented, too.

[adrfantini]

Still it wouldn't be bad to specify that the powersaving impact of those changes are probably minimal for SSDs. I would think that with a modern SSD those settings are probably best left at the default values as the benefits are very limited, and you pay a small price in responsiveness and amount of data lost on a crash. Thanks for the good job, by the way. Your tweaks are very comprehensive and useful. 2017-01-27 11:50 GMT+01:00 Jente Hidskes <[email protected]>:

…

That depends on what you mean with "not necessary", I suppose. If your goal is power saving, then you are probably correct in that it does not save as much power on an SSD as it does on an HDD (haven't tested how much of a difference, if any, it makes. I assume it would be negligible). However, less writing improves the lifespan of your SSD so in that sense you could deem it necessary (although, with modern SSDs, this is also not much of an issue anymore). In any case, it won't hurt to have them listed. Most of the entries in that file are currently commented, too. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#4 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AMO2HQzVh_xLdic1S-9uqmmUJebR2Bekks5rWcvjgaJpZM4Lvjp0> .

[Hjdskes]

Still it wouldn't be bad to specify that the powersaving impact of those
changes are probably minimal for SSDs.

I'm planning on writing an in-depth blog post about these settings, in which I will definitely mention this. I'll add it to the README too.

I would think that with a modern SSD those settings are probably best left at the default values as the benefits are very limited, and you pay a small price in responsiveness and amount of data lost on a crash.

They are left at the defaults, except for vm.dirty_writeback_centisecs (which, I admit, is a leftover from the time where I did not have an SSD - I'll probably change it when I write that blog post). In this case I'd still like to keep them in, if only to notify others of their existence. I have not noticed a (however small) decrease in responsiveness nor (increased) data loss on a system crash. In any case, feel free to change the settings to whatever you prefer!

[adrfantini]

Yeah, of course. Just wanted to make sure that it was clear that on SSDs the settings do not change much. All the best, thanks! 2017-01-28 15:44 GMT+01:00 Jente Hidskes <[email protected]>:

…

Still it wouldn't be bad to specify that the powersaving impact of those changes are probably minimal for SSDs. I'm planning on writing an in-depth blog post about these settings, in which I will definitely mention this. I'll add it to the README too. I would think that with a modern SSD those settings are probably best left at the default values as the benefits are very limited, and you pay a small price in responsiveness and amount of data lost on a crash. They *are* left at the defaults, except for vm.dirty_writeback_centisecs (which, I admit, is a leftover from the time where I did not have an SSD - I'll probably change it when I write that blog post). In this case I'd still like to keep them in, if only to notify others of their existence. I have not noticed a (however small) decrease in responsiveness nor (increased) data loss on a system crash. In any case, feel free to change the settings to whatever you prefer! — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#4 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AMO2HbvQlMrE-ahAvIo51dajztuqDV5Oks5rW1Q8gaJpZM4Lvjp0> .

[Hjdskes]

No problem, I appreciate the message!