On Sat, Jul 9, 2011 at 1:30 PM, iosif <iosif.neit...@gmail.com> wrote:
> Is it correct that the program rzip was not recommended because of its > lack of pipe support? By the Linux Journal article or by me? The lack of pipe support is pretty much a killer for the application that I was thinking of, which was copying a disk image via netcat to another machine over the network. But basically, I found the chart pretty hard to use, and I wish they had not logged the x-axis. There was one circle for rzip at around the 1 second mark, but there were like 5 of them around the 10 second mark. Because the axis is log scaled, left-right differences don't seem as extreme as they are. It's much more important to be near the left of the chart (i.e. fast) than it is to be near the top of the chart (i.e. space efficient) since the x-axis changes by magnitudes. That is what makes lzop stand out to me. So I did a little informal testing, and lzop seems pretty swell. I expect it would be good for something like this: machine 2: ---------- # cd /; netcat -l 4000 | tar --lzop xf - machine 1: ---------- # cd /; tar --one-file-system --lzop cf - | netcat -q 1 machine2 4000 Don
