hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From stack <saint....@gmail.com>
Subject Re: Doc Review: Accordion: In-memory Compaction
Date Fri, 07 Apr 2017 22:13:52 GMT
Yes sir.  Coming....

On Apr 7, 2017 10:24 AM, "Jean-Marc Spaggiari" <jean-marc@spaggiari.org>
wrote:

> Nice! Very interesting! Thanks for sharing!
>
> Are we also going to publish any numbers regarding the performance
> improvement?
>
> JMS
>
> 2017-04-07 12:39 GMT-04:00 Stack <stack@duboce.net>:
>
> > Devs:
> >
> > As a good few of you know already, our brother Edward and sisters
> > (Anastasia and Eschar) at Yahoo! Research, Israel -- with help from a
> good
> > few of you (Ram, Anoop, Duo, Ted, etc.) -- have been working furiously
> for
> > a good while now on a new feature for 2.0.0 that does In-memory flushes
> and
> > compaction. They are mostly done and are now finishing up the user-facing
> > [1] and dev-facing [2] documentation. We intend to post both views on the
> > Apache HBase blog.
> >
> > I am writing because, while our usual style is to just dump notice of the
> > posting into dev/user lists after review by those close-in, I'm writing
> > ahead of the posting petitioning review for two reasons:
> >
> > # There is the obvious, reviews will help improve the quality of the
> > posting.
> > # but, the main reason I am writing is that these samples are a "joy to
> > read", the dev view in particular. I find that it is of a quality I wish
> we
> > could emulate more frequently hereabouts: terse, clear, clean diagrams,
> and
> > well reasoned.
> >
> > If you get a chance, take a look.
> >
> > We'll probably post to the blog start of next week.
> >
> > Thanks,
> > St.Ack
> >
> > 1.
> > https://docs.google.com/document/d/1lsDv8mmw3Daz9Rw9zySEI7zXOlLYY
> > y2dyhQoB6gNMcI/edit#heading=h.7eh0oxf0l8yz
> > 2.
> > https://docs.google.com/document/d/1z1R-MdAxRvTC2NazxUmN3FOCFIknkxL2TF
> > qVUhYBVbE/edit#heading=h.5wx81z76o3nt
> >
>

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