2018-03-19 13:06 időpontban Sad Clouds ezt írta:
Hello, which virtual controller do you use in VirtualBox and do you have "Use Host I/O Cache" selected on that controller? If yes, then you need to disable it before running I/O tests, otherwise it caches loads of data in RAM instead of sending it to disk.On Mon, Mar 19, 2018 at 8:59 AM, Martin Husemann <martin%duskware.de@localhost> wrote:On Mon, Mar 19, 2018 at 08:54:12AM +0000, Chavdar Ivanov wrote:I'd be also interested in your setup - on a W10 hosted VBox (latest) on a fast M.2 disk I get approximately 5 times slower values, on -current amd64, having disks attached to SATA, SAS and NVMe controllers (almost the same,the SAS one is a little slower than the rest, but nowhere near your figures. :Hmm, nothing special, latest VBox, Win7 host, plain old hard disk asbackend store with NTFS on a SATA disk. But the host has *plenty* of memory,maybe I should have used a larger dd to exhaust buffering. Martin
Hi There, I've performed a short test. Hereby my results: HOST: Win7, Intel Core5, 2 CPU's 1. VM: NetBSD 7.1.1, original kernel Controller: SATA Driver: AHCI FS: ffs, default settings Command: dd if=/dev/zero of=out bs=1m count=1000 Host I/O cache off. Average result of 3 runs: 105,5 MB/sec 2. VM: Debian 9.0, original kernel Controller: SATA Driver: AHCI FS: ext4, default settings Command: dd if=/dev/zero of=out bs=1M count=1000 Host I/O cache off. Average result of 3 runs: 588,0 MB/secSo, Debian performed almost 6 times faster than NetBSD on the same machine.
Any setting which influence the test and I didn't apply? Rgds, FeZ