You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@cloudstack.apache.org by Sebastien Goasguen <ru...@gmail.com> on 2015/04/06 21:59:22 UTC

Re: Review Request 20184: CLOUDSTACK-6369: Qcow2's cluster_size option support

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/20184/#review78999
-----------------------------------------------------------


Thank you for submitting your CloudStack contribution through review board. After discussion on the dev@cloudstack.apache.org the community decided to close down review board and start accepting contributiong through GitHub pull requests. We have been using GH PR for several months now and the process is better than review board.

We will keep Review Board open for another week to give you time to migrate your patch to a github PR if you wish. After that time, your patch will no longer be viewable (even though it will not be deleted).

Please consider submitting a pull request.

Great instructions are available at:
https://github.com/apache/cloudstack/blob/master/CONTRIBUTING.md

Thank you very much for your time and your contribution to Apache CloudStack, we hope that using this new process will encourage you to do more.

- Sebastien Goasguen


On April 9, 2014, 11:21 p.m., Yoshikazu Nojima wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/20184/
> -----------------------------------------------------------
> 
> (Updated April 9, 2014, 11:21 p.m.)
> 
> 
> Review request for cloudstack, Mike Tutkowski, Marcus Sorensen, and Wido den Hollander.
> 
> 
> Bugs: CLOUDSTACK-6369
>     https://issues.apache.org/jira/browse/CLOUDSTACK-6369
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> -------
> 
> To improve the qcow2 performance, it is important to tune parameters of "qemu-img create" command.
> preallocation=metadata option, which I am working on in CLOUDSTACK-6191 improves randam IOPS well, but the qcow2 image created with preallocation=metadata option cannot be used as a differential image.
> This "cluster_size" option improves IOPS of a differential qcow2 image which is created with preallocation=off.
> 
> https://issues.apache.org/jira/browse/CLOUDSTACK-6369
> 
> This patch is written on the top of https://reviews.apache.org/r/19446/ . This patch cannot be applied before 19446 is merged.
> 
> 
> Diffs
> -----
> 
>   agent/conf/agent.properties b8b7a7c 
>   core/src/com/cloud/storage/JavaStorageLayer.java e8c4053 
>   plugins/hypervisors/kvm/src/com/cloud/hypervisor/kvm/storage/LibvirtStorageAdaptor.java 61d90e9 
> 
> Diff: https://reviews.apache.org/r/20184/diff/
> 
> 
> Testing
> -------
> 
> Case1 (without this option)
> 1. Ensure "storage.qcow2.cluster_size" is not written in "agent.properties" file.
> 2. Create and attach a volume.
> 3. Confirm the created volume is provisioned with cluster_size=65536, which is a default value.
> 
> Case2 (with this option)
> 1. Ensure "storage.qcow2.cluster_size=2097152" is written in "agent.properties" file.
> 2. Create and attache a volume.
> 3. Confirm the created volume is provisioned with cluster_size=2097152.
> 
> 
> Thanks,
> 
> Yoshikazu Nojima
> 
>