Anupama Gupta has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/11263 )

Change subject: Blogpost describing index skip scan optimization.
......................................................................


Patch Set 7:

(13 comments)

Many thanks for the comments. Please take a look.

http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md
File _posts/2018-08-17-index-skip-scan-optimization-in-kudu.md:

http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@39
PS6, Line 39: table
> nit: tablet, here and elsewhere
Done


http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@45
PS6, Line 45: (we will refer to it as
            : "prefix column" and its specific value as "prefix key").
> nit: since we're not using these as a variable names, but rather as definit
Done


http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@48
PS6, Line 48: Therefore, we can use the index to skip to the rows that have 
distinct prefix keys,
            : and also satisfy the predicate on the `tstamp` column.
> nit: maybe drop the ** around "skip" here, since you do it down below anywa
Done


http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@58
PS6, Line 58: `host = helium`
> nit: would be nice if the entire thing were in backticks, since it's a cond
You are right. Made the change.


http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@60
PS6, Line 60: satisfy the predicate, and we
> nit: probably not needed
Done


http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@59
PS6, Line 59: . At that
            : point we would
> nit: reword "until the predicate no longer matches. At that point we would
Done


http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@67
PS6, Line 67: tio
> nit: this is a little distracting, below too. Let's just keep it singular s
Done


http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@73
PS6, Line 73: o get worse with respect to the full tablet scan performance when 
the prefix column cardinality
> This and below are being rendered weirdly by github. Would like to confirm
Yes, it works fine with jekyll. (Link to this screen shot - 
https://raw.githubusercontent.com/AnupamaGupta01/kudu-1/gh-pages-staging/img/index-skip-scan/equation.png)


http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@74
PS6, Line 74: C%20tablet%20%7D).
            : Therefore, in order to use skip scan perf
> "consistent performance in cases of large prefix column cardinality"
Done


http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@93
PS6, Line 93:
> nit: probably not needed, below too.
Done


http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@94
PS6, Line 94: Range pr
> nit: In-list
Done


http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@98
PS6, Line 98: orki
> nit: team
Done


http://gerrit.cloudera.org:8080/#/c/11263/6/_posts/2018-08-17-index-skip-scan-optimization-in-kudu.md@101
PS6, Line 101:
             : References
             : ==========
             :
             : 
[[1]](https://storage.googleapis.com/pub-tools-public-publication-data/pdf/42851.pdf):
 Gupta, Ashish, et al. "Mesa:
             : Geo-replicated, near real-time, scalable data warehousing." 
Proceedings of the VLDB Endowment 7.12 (2014): 1259-1270.
             :
             : [[2]](https://oracle-base.com/articles/9i/index-skip-scanning/): 
Index Skip Scanning - Oracle Database
             :
> It's really up to you, but WDYT about just linking these in-line? This is a
Thanks, I see your point. I think that the current section for references looks 
fine after incorporating Alexey's suggestions on the same (in Patch 4, L62).



--
To view, visit http://gerrit.cloudera.org:8080/11263
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: gh-pages
Gerrit-MessageType: comment
Gerrit-Change-Id: I2250652dcba3d1b0a06f1ffb7f23c11bf533d35e
Gerrit-Change-Number: 11263
Gerrit-PatchSet: 7
Gerrit-Owner: Anupama Gupta <[email protected]>
Gerrit-Reviewer: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Andrew Wong <[email protected]>
Gerrit-Reviewer: Anupama Gupta <[email protected]>
Gerrit-Reviewer: Attila Bukor <[email protected]>
Gerrit-Reviewer: Mike Percy <[email protected]>
Gerrit-Comment-Date: Thu, 13 Sep 2018 03:35:44 +0000
Gerrit-HasComments: Yes

Reply via email to