SSD write exhaustion is massively overstated for normal consumers.
At 12GB/day, the shortest-lived drive that Techreport endurance tested to death would last 187 years. I’m not really worried about the life of my drives. I’m still concerned that Firefox is so incredibly unparsimonious, but welcome to the world of the modern Web.
A dozen gigabytes here, a dozen gigabytes there, and pretty soon we’re talking real numbers.
Sure, if I somehow manage to write twenty dozen gigabytes every day, my SSD will die from write exhaustion after… nine years.
Nine years ago, a typical hard drive stored on the order of 200GB, an amount I can now buy on a USB thumb drive for less than $50. I don’t think I will lament the loss of this hypothetical SSD.
Don’t get me wrong, Firefox’s write habits are concerning, just not because they will kill SSDs.
I’m not particularly worried, but did you do the math for those of us with only a 32GB SSD that’s already mostly full, so there’s not much free space left to wear level over?
My numbers aren’t quite so bad, but I’ve noticed if I have two blank tabs open, and open a third blank tab, that requires saving 300K of state. Which seems like a lot of state to record practically nothing.
It’s a shame that KHTML was subsumed and supplanted by Webkit. The implementation was clean, simple, and solid. It seems more aligned with the goals a lot of us crustaceans have.
My experience was different. I use OpenBSD as my desktop and noticed that after a while Firefox would just get slow and peg the cpu as much as possible–with a significant portion of cpu going to system. I eventually tracked it down to disk I/O–about 1ktps and about 16MBps. I tried moving my profile onto an mfs mount, but it still showed excessive CPU usage. The fix for me? Turn e10s off. I was excited about it and had turned it on. Things are much better now.
I’ve been using SSDs with FireFox for years and none of them has died.