cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Stefania (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-8894) Our default buffer size for (uncompressed) buffered reads should be smaller, and based on the expected record size
Date Wed, 08 Jul 2015 01:05:04 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-8894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14617778#comment-14617778
] 

Stefania commented on CASSANDRA-8894:
-------------------------------------

Yes I assumed a normal distribution of the record size. Your suggestion of a uniform distribution
of _start position_ within a page is more straight-forward however. Let's start with that:
{{size}} = 95 percentile, chance of crossing = {{(size % 4096) / 4096}}

Noted about adding size percentile and chance of crossing threshold to the config without
mention in the yaml. I'll also add a *global* setting to indicate if the data directories
are SSD or spinning disk, and this will instead be in the yaml. 

> Our default buffer size for (uncompressed) buffered reads should be smaller, and based
on the expected record size
> ------------------------------------------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-8894
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-8894
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>            Reporter: Benedict
>            Assignee: Stefania
>              Labels: benedict-to-commit
>             Fix For: 3.x
>
>
> A large contributor to slower buffered reads than mmapped is likely that we read a full
64Kb at once, when average record sizes may be as low as 140 bytes on our stress tests. The
TLB has only 128 entries on a modern core, and each read will touch 32 of these, meaning we
are unlikely to almost ever be hitting the TLB, and will be incurring at least 30 unnecessary
misses each time (as well as the other costs of larger than necessary accesses). When working
with an SSD there is little to no benefit reading more than 4Kb at once, and in either case
reading more data than we need is wasteful. So, I propose selecting a buffer size that is
the next larger power of 2 than our average record size (with a minimum of 4Kb), so that we
expect to read in one operation. I also propose that we create a pool of these buffers up-front,
and that we ensure they are all exactly aligned to a virtual page, so that the source and
target operations each touch exactly one virtual page per 4Kb of expected record size.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message