Messages in this thread | | | Date | Tue, 22 Dec 1998 13:25:28 -0500 | From | (Chuck Lever) | Subject | Re: Swap Cache |
| |
> From: "Paul R. Wilson" <wilson@cs.utexas.edu> > Date: Mon, 21 Dec 1998 17:28:11 -0600 > Subject: Re: Swap Cache > > >While your idea is correct, you seem to be forgetting > >the fact that disk transfers can be done in parallel > >with useful CPU operations in other tasks. > > I do realize that it's an issue. I just don't believe > it's fatal. For many workloads, a system is either CPU bound > or I/O bound, and you can easily tell which. And as CPU's > get faster and memory gets larger, the tradeoffs get less > tense.
i think that paging and CPU operations only happen at the same time on a system that has many things to do. in most cases the CPU must wait for page operations to complete anyway. but that point has probably already been made.
> I have a bit of a feeling that compressed caching is being > criticized on shifting grounds. If RAM is cheap, then > buy a little extra and use compression to double it, so > that you never page. If RAM is expensive, and you're > I/O bound, then you clearly need it. Take your pick.
seems to me putting more than 4G into an Intel PC is "infinitely expensive" -- for anything other than a Xeon CPU, you can't do it. does anyone think that using page compression might help ease this constraint? what if this were a way to effectively use more than 4G by, say, making the "high memory" area a swap staging area?
in other words, this type of effort may have more wide use than just as a way to make a poor student's machine more cost effective.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |