kafka-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jay Kreps <jay.kr...@gmail.com>
Subject Re: Database Replication Question
Date Wed, 11 Mar 2015 15:37:48 GMT
Xiao,

Not sure about AIX or HP-UX. There are some people running on Windows,
though we don't do real systemic testing against that. I would be surprised
if z/os worked, someone would have to try.

The existing fsync policy already works at the batch level, and Kafka
already does batching quite aggressively. If you set the fsync policy to
force fsync and you get an acknowledgement that the fsync occurred.

In any case, it has not been my experience that datacenter power outages
are a common failure mode whereas disk failures happen continually.

I agree that the prototype transaction support would not be a good fit for
trying to commit each message in a transaction.

-Jay



On Mon, Mar 9, 2015 at 10:56 PM, Xiao <lixiao1983@gmail.com> wrote:

> Hi, Jay,
>
> Thank you!
>
> The Kafka document shows “Kafka should run well on any unix system". I
> assume it includes the major two Unix versions, IBM AIX and HP-UX. Right?
>
> 1. Unfortunately, we aims at supporting all the platforms, Linux, Unix,
> Windows and especially z/OS. I know z/OS is not easy to support.
>
> 2. Fsync per message is very expensive and Fsync per batch will break the
> transaction atomicity. We are looking for transaction-level fsync, which is
> more efficient. Then, our producers can easily combine multiple small
> transactions into a single bigger batch transaction. I hope the
> implementation of the ongoing Kafka feature “Transactional Messaging in
> Kafka” already considered all these issues, although the following link
> does not mention it:
> https://cwiki.apache.org/confluence/display/KAFKA/Transactional+Messaging+in+Kafka
>
> 3. After a quick reading of the design of “Transactional Messaging in
> Kafka”, I have a doubt if it can scale especially when most transactions
> are very short (e.g., containing a single message).
>         - In many use cases, we do not need global transactions, which
> might be too expensive. The partition-specific transaction granularity
> might be fine.
>         - To reduce the overhead of transaction-level fsync + network or
> address-space roundtrips, Kafka might need two extra parameters for
> batching the small transactions into a single one. The performance benefits
> of batching are huge, as shown in the MRI (multi-row insert) feature in
> Oracle and DB2 z/OS.
>
> I believe transactional messaging is a critical feature. The design
> document is not very clear. Do you have more materials or links about it?
>
> Thanks,
>
> Xiao Li
>
>
> On Mar 7, 2015, at 9:33 AM, Jay Kreps <jay.kreps@gmail.com> wrote:
>
> > Xiao,
> >
> > FileChannel.force is fsync on unix.
> >
> > To force fsync on every message:
> > log.flush.interval.messages=1
> >
> > You are looking at the time based fsync, which, naturally, as you say, is
> > time-based.
> >
> > -Jay
> >
> > On Fri, Mar 6, 2015 at 11:35 PM, Xiao <lixiao1983@gmail.com> wrote:
> >
> >> Hi, Jay,
> >>
> >> Thank you for your answer.
> >>
> >> Sorry, I still do not understand your meaning.
> >>
> >> I guess the two parameters you mentioned are log.flush.interval and
> >> log.default.flush.interval.ms. However, these two parameters only
> control
> >> when Kafka issues a flush (i.e., calling FileChannel.force()).
> >>
> >> Fsync (fileOutputStream.getFD().sync()) is controlled by another
> parameter
> >> log.default.flush.scheduler.interval.ms.
> >>
> >>      scheduler.schedule("kafka-recovery-point-checkpoint",
> >>                         checkpointRecoveryPointOffsets,
> >>                         delay = InitialTaskDelayMs,
> >>                         period = flushCheckpointMs,
> >>                         TimeUnit.MILLISECONDS)
> >>
> >> This thread is only time-controlled. It does not check the number of
> >> messages.
> >>
> >> Thank you,
> >>
> >> Xiao Li
> >>
> >>
> >> On Mar 5, 2015, at 11:59 AM, Jay Kreps <jay.kreps@gmail.com> wrote:
> >>
> >>> Hey Xiao,
> >>>
> >>> That's not quite right. Fsync is controlled by either a time based
> >> criteria
> >>> (flush every 30 seconds) or a number of messages criteria. So if you
> set
> >>> the number of messages to 1 the flush is synchronous with the write,
> >> which
> >>> I think is what you are looking for.
> >>>
> >>> -Jay
> >>
> >>
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message