Red Hat Bugzilla – Bug 633380
[6.2 FEAT] Include QED image format for KVM guests
Last modified: 2014-07-24 23:07:00 EDT
1. Feature Overview: Feature Id: [67243] a. Name of Feature: [6.1 FEAT] Include QED image format for KVM guests b. Feature Description QED is a disk image format that supports sparse images and backing files at new levels of performance and data integrity which are not available in existing formats. It also provides a base for improved block migration/streaming (see separately filed features). 2. Feature Details: Sponsor: LTC Open Virtualization Architectures: x86_64, Arch Specificity: purely common code Affects Kernel Modules: Field does not exist Delivery Mechanism: Direct from Community Category: other Request Type: Package - Update Version d. Upstream Acceptance: Field does not exist Sponsor Priority P3 f. Severity: normal IBM Confidential: No Code Contribution: IBM code g. Component Version Target:--- 3. Business Case A high performance and reliable disk format that supports fast provisioning is needed for the IBM public cloud to reduce the time for VM deployment. 4. Primary contact at Red Hat: John Jarvis jjarvis@redhat.com 5. Primary contacts at Partner: Project Management Contact: Michael W. Wortman, wortman@us.ibm.com Technical contact(s): Stefan Hajnoczi, stefan.hajnoczi@uk.ibm.com
What is the component for this and is it upstream? Please provide more details about is being requested here.
------- Comment From ryanh@us.ibm.com 2010-09-21 17:28 EDT------- Sorry, our BZ doesn't have much in the component bucket for userspace changes. This is a patchset to qemu. It isn't upstream yet; RFC posted to community, mergable patches coming out this week. ------- Comment From aliguori@us.ibm.com 2010-09-21 17:30 EDT------- The component is QEMU, the patches have been posted and are expected to be accepted upstream shortly. We don't have the ability to select a component in our BZ strangely enough.
Great, thanks.
------- Comment From rsisk@us.ibm.com 2010-10-04 11:22 EDT------- Code Upstream Status: Submitted
We need to be fully convinced that the QED perf/functionality cannot be provided with qcow2. That's why for the moment it's in conditional_nak state by design.
After long evaluation period we came to conclusion that it's too early to include QED in rhel6.1 and we rather try improve qcow2/qcow3 in order to get to similar performance figures.
------- Comment From ryanh@us.ibm.com 2011-02-04 18:22 EDT------- The IBM Public Compute Cloud requires a image format that has both strong data integrity as well as very high (raw) performance that can be used to eliminate image deployment time. This capability has a direct impact on Compute Cloud revenue generation and is a critical, must-have feature. Currently qcow2 does not meet the above requirements w.r.t performance and data integrity. The deficiencies have been discussed in great detail in technical interlocks in the community and between IBM and RedHat. Furthermore, the current planning around qcow3 is not a sufficient alternative due to the time-frame in which this feature is needed for the IBM Compute Cloud business. The IBM Public Compute Cloud is scheduled to deploy this feature in 2011 based on RHEL6.1. The requested feature is already available and committed upstream in QEMU. IBM is requesting this upstream feature be included as Tech Preview in RHEL6.1.
------- Comment From tpnoonan@us.ibm.com 2011-02-05 13:44 EDT------- (In reply to comment #12) > The IBM Public Compute Cloud requires a image format that has both strong data > integrity as well as very high (raw) performance that can be used to eliminate > image deployment time. This capability has a direct impact on Compute Cloud > revenue generation and is a critical, must-have feature. Currently qcow2 does > not meet the above requirements w.r.t performance and data integrity. The > deficiencies have been discussed in great detail in technical interlocks in the > community and between IBM and RedHat. Furthermore, the current planning around > qcow3 is not a sufficient alternative due to the time-frame in which this > feature is needed for the IBM Compute Cloud business. > The IBM Public Compute Cloud is scheduled to deploy this feature in 2011 based > on RHEL6.1. > The requested feature is already available and committed upstream in QEMU. > IBM is requesting this upstream feature be included as Tech Preview in RHEL6.1. Bottom line QEMU delivers the required provisioning performance while qcow2 does not.
Hi Anthony, QED is mainly developed and tested by IBM, can IBM test it? Thanks Suqin
IBM is signed up to test and provide feedback, setting OtherQA.
This enhancement request was evaluated by the full Red Hat Enterprise Linux team for inclusion in a Red Hat Enterprise Linux minor release. As a result of this evaluation, Red Hat has tentatively approved inclusion of this feature in the next Red Hat Enterprise Linux Update minor release. While it is a goal to include this enhancement in the next minor release of Red Hat Enterprise Linux, the enhancement is not yet committed for inclusion in the next minor release pending the next phase of actual code integration and successful Red Hat and partner testing.
------- Comment From stefan.hajnoczi@uk.ibm.com 2011-10-04 08:34 EDT------- I have verified this feature with: * RHEL 6.2 guest install * Fedora 16 LXDE guest install * qemu-iotests
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHSA-2011-1531.html