In my opinion it pushes the boundaries of what is appropriate for a
self-review, since this represents a potentially significant new
behavior that will be exposed to customers and ISVs.
However, it also looks fairly obvious and straight-forward.
Can I kindly request that you promote this to a fast track, and accept
my +1 at the same time?
(Unless someone else has some significant concern, the only negative
implication of this is that you'll need to wait a few days before
integrating ... probably this will be approved on Wednesday.)
- Garrett
Tim Haley wrote:
> I am sponsoring the following case on behalf of Neil Perrin.
> Requested binding is Micro/Patch. The case is a straight-forward
> addition of a new property on zfs datasets for tuning separate
> intent log usage. I believe the case qualifies for self review and
> have filed it as closed approved automatic. If folks disagree, let me
> know and I'll convert it to a fast-track.
>
> Template Version: @(#)sac_nextcase 1.68 02/23/09 SMI
> This information is Copyright 2009 Sun Microsystems
> 1. Introduction
> 1.1. Project/Component Working Name:
> ZFS logbias property
> 1.2. Name of Document Author/Supplier:
> Author: Neil Perrin
> 1.3 Date of This Document:
> 07 August, 2009
>
> 4. Technical Description
>
> Summary
>
> Provide zfs with the ability to control the use of resources
> used for synchronous (eg fsync, O_DSYNC) requests. In particular
> it enables substantially better performance for Oracle and
> potentially other applications.
>
> Background
>
> Oracle manages two major types of files, the Data Files and the Redo Log
> files. Writes to Redo Log files are in the path of all transactions
> and low latency is a requirement. It is critical for good performance
> to give high priority to these write requests.
>
> Data Files are also the subject of writes from DB writers as a form of
> scrubbing dirty blocks to insure DB buffer availability. Write latency is
> much less an issue. Of more more importance is achieving an acceptable level
> of throughput. These writes are less critical to delivered performance.
>
> Both types of writes are synchronous (using O_DSYNC), and thus treated
> equally. They compete for the same resources: separate intent log,
> memory, and normal pool IO. The Data File writes impede the potential
> performance of the critical Redo Log writers.
>
> Proposal
>
> Create a new "logbias" property for zfs datasets.
>
> If logbias is set to 'latency' (the default) then there is no change from the
> current implementation. If the logbias property is set to 'throughput'
> then intent log blocks will be allocated from the main pool instead of any
> separate intent log devices (if present). Also data will be written
> immediately to spread the write load thus making for quicker subsequent
> transaction group commits to the pool.
>
> To change the property, an admin will use the standard zfs set command:
>
> # zfs set logbias=latency {dataset}
> # zfs set logbias=throughput {dataset}
>
> The current value of logbias can be retrieved in the normal
> manner with 'zfs get logbias' or 'zfs list -o logbias'.
>
> Manpage diffs
>
> --- old=09Thu May 7 15:41:08 2009
> +++ new=09Thu May 7 15:44:30 2009
> @@ -1038,7 +1038,15 @@
> perty was changed. If the new property is "off", the
> file systems are unshared.
>
> + logbias = latency | throughput
> +
> + Provide a hint to ZFS about handling of synchronous
> + requests in this dataset. If logging is set to "latency" (the
> + default) ZFS will use pool log devices (if configured) to handle the
> + requests at low latency. If logging is set to "throughput" then ZFS
> + will not use configured pool log devices. ZFS will
> + instead optimize synchronous operations for global
> + pool throughput and efficient use of resources.
>
> 6. Resources and Schedule
> 6.4. Steering Committee requested information
> 6.4.1. Consolidation C-team Name:
> ON
> 6.5. ARC review type: Automatic
> 6.6. ARC Exposure: open
>
>
