<p dir="ltr">Congratulations!</p>
<div class="gmail_quote">On Jun 3, 2013 12:13 AM, "MORITA Kazutaka" <<a href="mailto:morita.kazutaka@gmail.com">morita.kazutaka@gmail.com</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hi all,<br>
<br>
I'm pleased to announce the availability of the 0.6.0 release.<br>
The software is available from GitHub:<br>
<a href="https://github.com/collie/sheepdog/archive/v0.6.0.tar.gz" target="_blank">https://github.com/collie/sheepdog/archive/v0.6.0.tar.gz</a><br>
<br>
Major features since 0.5.6 include:<br>
<br>
- Add MD support<br>
<br>
Sheepdog can manage multiple disks with one daemon now. The basic<br>
idea of MD is RAID-0 like mechanism that distributes sheep objects<br>
on the local disks without parity or replicating, which instead<br>
relies on the sheepdog's replicated storage to recover the lost<br>
objects on the faulty disk.<br>
<br>
- Add journaling support<br>
<br>
Yuan reworked the old journal implementation and boosted random<br>
write performance of sheep.<br>
<br>
- Move farm from sheep to collie<br>
<br>
The farm cluster snapshot in sheep had some drawbacks:<br>
<br>
1. After a new node joins the cluster, all former snapshots are<br>
not available to be restored.<br>
2. It is hard to backup cluster snapshot to another storage<br>
system for disaster-tolerance.<br>
3. It is hard to initialize a new cluster by loading another<br>
cluster's snapshot.<br>
<br>
New implementation is to move "farm" from sheep to collie and save<br>
cluster snapshot to a local path. This retains all features from<br>
"farm", including:<br>
<br>
1. object de-duplication<br>
2. incremental store capability<br>
<br>
In addition, it also provides ability of:<br>
<br>
1. export cluster snapshot to other storage device for backup and<br>
disaster-tolerance<br>
2. deploy new cluster by restore from one snapshot of other<br>
cluster<br>
<br>
- Use hash check for vdi check and recovery<br>
<br>
To reduce network traffic, Sheepdog uses a hash values to compare<br>
objects between local and remote.<br>
<br>
- Shepherd<br>
<br>
Currently, corosync doesn't have enough scalability and ZooKeeper<br>
is not stable under heavy network traffic. This is an initial<br>
implementation for yet another cluster coordinator to solve the<br>
problem.<br>
<br>
Shepherd is not mature yet. Don't use it for other than testing<br>
purposes.<br>
<br>
- Add support for a trim command<br>
<br>
With this support, Sheepdog can remove the objects which are<br>
unnecessary for the guest virtual machines.<br>
<br>
<br>
The next release 0.7.0 will be in the end of August. We decided to<br>
use the same release cycle as QEMU. I'll announce our schedule after<br>
QEMU community announce theirs.<br>
<br>
Thanks,<br>
<br>
Kazutaka<br>
--<br>
sheepdog mailing list<br>
<a href="mailto:sheepdog@lists.wpkg.org">sheepdog@lists.wpkg.org</a><br>
<a href="http://lists.wpkg.org/mailman/listinfo/sheepdog" target="_blank">http://lists.wpkg.org/mailman/listinfo/sheepdog</a><br>
</blockquote></div>