3 ms·
You’re not correct on a number of fronts here. ZFS is not tied to the memory allocator in Solaris, I’m very unclear how you could have made that mistake. 128b
by dijit 5y ago
You’re not correct on a number of fronts here.
ZFS is not tied to the memory allocator in Solaris, I’m very unclear how you could have made that mistake.
128bits might feel wasteful, but you’re likely never going to have the problems FAT32 has had- it’s designed for the next 100 years- which has drawbacks but prevent the kinds of ugly hacks that NTFS has gone through to maintain forward compatibility (rolling on disk upgrade).
And yes, the 1G per TB is for the in memory block map, which is a requirement for deduplucation on the default block size.
- jcalvinowens 5y ago> ZFS is not tied to the memory allocator in Solaris, I’m very unclear how you could have made that mistake > And yes, the 1G per TB is for the in memory block map, which is a requirement for deduplucation on the default block size. You actually answered your own question: it's antiquated because of the assumption it is possible to get contiguous RAM for that mapping. On modern Linux, that is only possible if you reserve it pre-boot.
- dijit 5y agoThe Solaris memory allocator is not contiguous either though
- jcalvinowens 5y agoThis is a whole rabbit hole... but essentially, Linux uses a naive buddy allocator for physical RAM, and Solaris was more interesting.
- detaro 5y ago> it's antiquated because of the assumption it is possible to get contiguous RAM for that mapping. source on that? (and even if that's the case, wouldn't that be an implementation detail? E.g. implementations can spill it to L2ARC, but somehow not to different memory locations?)
- jcalvinowens 5y agoMy memory is that this mapping does have to be contiguous, but I'll dig it up in the code and link it here either way.
- culpable_pickle 5y agostill planning on linking that code?