cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tyler Hobbs (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, 24 Jun 2015 15:56:05 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-8894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14599619#comment-14599619
] 

Tyler Hobbs commented on CASSANDRA-8894:
----------------------------------------

FYI, {{estimatedRowSize}} should really be called {{estimatedPartitionSize}} -- it's referring
to Thrift rows, not CQL rows.  When there are no clustering columns, there's only one row
per partition, so you could use that metric.  Otherwise, you need to try to deduce the average
number of cells per CQL row (and the number of cells per partition, but there's a metric for
that).  The number of cells per CQL row is pretty predictable in most cases, but collections
really throw that off, so I'm not sure what to do there.

> 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
>             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