[Users] Big comparison Ploop and ext4 vs simfs vs ZFS

Hello, folks! Big thanks for your attention to my work about ZFS: https://github.com/pavel-odintsov/OpenVZ_ZFS And I would like to share with community our own ZFS tests from FastVPS Eesti OU experience (fastvps.ru). We are company with 5 year experience with OpenVZ in heavy production usage. W

Re: [Users] ZFS vs ploop

It is not normal deduplication with FS tools. And we said about delete data and problem with TRIM it. If we do not remove anything and never - it is not bad. But in real life we create it with vzctl create, without ln. In real life we write and DELETE data. And we believe that numbers that we see o

Re: [Users] ZFS vs ploop

On 23.07.2015 5:44, Kir Kolyshkin wrote: My experience with ploop: DISKSPACE limited to 256 GiB, real data used inside container was near 40-50% of limit 256 GiB, but ploop image is lot bigger, it use near 256 GiB of space at hardware node. Overhead ~ 50-60% I found workaround for this: run "/

Re: [Users] SIMFS users

On 23.07.2015 3:45, Scott Dowdle wrote: vzctl has a compact option that will basically take the free space and give it back to the host. I've used compact a few times but I don't use it regularly... so I'm not sure how efficient it is nor how good it is at reclaiming 100% of the unused space...

Re: [Users] ZFS vs ploop

Completely agree with Gena Makhomed on points he raised about ploop. If you run HN on SSD (256gb for example) ploop is not good to use at all, too much overhead of space. Would be nice to have better free space management in ploop somehow. Also about OpenVZ restore option: Here is real example

Re: [Users] Big comparison Ploop and ext4 vs simfs vs ZFS

Pavel, Thanks for sharing. I have a few comments. 1. The article only compares disk space usage and nothing else. Performance, memory footprint, maintainability etc are all left aside. If this is the only optimization criteria for you, it's fine, but usually such comparison won't work. Say, you c

Re: [Users] Big comparison Ploop and ext4 vs simfs vs ZFS

Greetings, - Original Message - > So, we could share this article: > http://www.stableit.ru/2015/07/effectiveness-of-zfs-usage-for-openvz.html > which shown how ZFS bit ext4 and ploop. >From your article written in red: "That's why we have only single file system which ready for 21 centu

Re: [Users] Show stopper while trying out Openvz 7 beta 1

Hello, Following up, it appears that my only problem was that one now must create an OS template cache, as in PVC, and the documentation on the wiki did not yet reflect that fact - but once the template cache is created, everything falls into place. The old precreated templates are apparently not

Re: [Users] Big comparison Ploop and ext4 vs simfs vs ZFS

> The point is that in your use case ZFS works very well and saves you 1TB of > disk space. Great for you. For my use case the overhead of learning and > deploying ZFS greatly outweighs the handful of GBs I might potentially save in > disk space. Don't get me wrong, but zfs is totally simply to

Re: [Users] Big comparison Ploop and ext4 vs simfs vs ZFS

Hello! Thanks for feedback! We have more details in Git repository page. But actually we haven't performance tests for this benchmarks because they are executed few months ago ;) On Friday, July 24, 2015, Kirill Kolyshkin wrote: > Pavel, > > Thanks for sharing. I have a few comments. > > 1. Th

Re: [Users] Big comparison Ploop and ext4 vs simfs vs ZFS

Hello! Completely agree with you. With old style filesystems we use so much non uniform crap like mdraid, dm raid and hardware solutions like lsi and adaptec. There are so much different tools, so much details and really complex configurations. With ZFS we have only singletool (ring!) for ewery

Re: [Users] ZFS vs ploop

On 07/24/2015 05:41 AM, Gena Makhomed wrote: To anyone reading this, there are a few things here worth noting. a. Such overhead is caused by three things: 1. creating then removing data (vzctl compact takes care of that) 2. filesystem fragmentation (we have some experimental patches to ext4

Re: [Users] ZFS vs ploop

On 25.07.2015 1:06, Kir Kolyshkin wrote: I think this is not good idea run ploop compaction more frequently, then one time per day at the night - so we need take into account not minimal value of overhead, but maximal one, after 24 hours of container work in normal mode - to planning disk space

[Users] Kernel 3.10 packages availability

I see there is a 3.10 kernel branch here: https://src.openvz.org/projects/OVZ/repos/vzkernel/browse Is it considered stable? Is there any plan to make packages (e.g. for Debian distributions) ? Would be nice to have an updated kernel for OpenVZ packaged automatically and get updates when you pus

Re: [Users] ZFS vs ploop

On 07/24/2015 05:57 PM, Gena Makhomed wrote: On 25.07.2015 1:06, Kir Kolyshkin wrote: what I am doing wrong, and how I can decrease ploop overhead here? Most probably it's because of filesystem defragmentation (my item #2 above). We are currently working on that. For example, see this report:

Re: [Users] Kernel 3.10 packages availability

On 07/24/2015 06:38 PM, spameden wrote: I see there is a 3.10 kernel branch here: https://src.openvz.org/projects/OVZ/repos/vzkernel/browse Is it considered stable? Looks like you missed all the hype about VZ7: https://lists.openvz.org/pipermail/announce/2015-April/000579.html In short --

Re: [Users] Kernel 3.10 packages availability

2015-07-25 5:07 GMT+03:00 Kir Kolyshkin : > > > On 07/24/2015 06:38 PM, spameden wrote: > > I see there is a 3.10 kernel branch here: > > https://src.openvz.org/projects/OVZ/repos/vzkernel/browse > > Is it considered stable? > > > Looks