> > I've stumbled across some mentions about this, but > no real details. > > What's this project about and what are the grand > plans, if you're > > allowed to disclose them? > > VM2 is a project to redesign the Solaris virtual > memory system around > modern computer architectures. The core of the > current VM system was > designed in 1985-86, when Sun's large computers had 4 > megabytes of RAM, > one simple CPU with a simple MMU, very few disks, and > no NUMA or power > management. A couple decades of Moore's Law and many > many billions of > dollars of hardware development later, things look a > bit different. > Obviously the software has evolved to deal with these > hardware changes, > and it's a testament to the original design that it's > performed for this > long. But the VM system has developed a reputation > for being hard to > understand, as 20 years' of accreted development will > tend to do, and > having most of the VM interfaces operate on lists of > small fixed-size > pages has made it hard to do more significant > innovation. > > We have a small team working on this project right > now, and we've been > focusing our limited time and energy on making sure > that our core > concepts are going to work as well as we expect. > Once our ideas are > ore solidified and tested, we'll write up a more > complete description.
Would it be reasonable to hope that this would also result in the elimination of the distinction between the page cache used by most filesystems and the ARC used by zfs (and thus presumably the problems getting best overall use of memory given how large the latter can grow if not explicitly limited by otherwise evil manual tuning)? Any broad idea of the degree to which this would impact drivers that in some way or another ultimately support mmap()? This message posted from opensolaris.org _______________________________________________ perf-discuss mailing list perf-discuss@opensolaris.org
