[
https://issues.apache.org/jira/browse/CASSANDRA-10331?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15188861#comment-15188861
]
Stefania edited comment on CASSANDRA-10331 at 3/14/16 3:02 AM:
---------------------------------------------------------------
In terms of establishing the canonical bulk reading workload, I suggest using
the profiles available in [this
benchmark|https://github.com/stef1927/cstar_bulk_read_test] with single
clustering and row sizes of approximately 100 bytes, 1 kb and 10 kb. These
correspond to the following user profiles:
*
[bulk_read_100_bytes.yaml|https://github.com/stef1927/cstar_bulk_read_test/blob/master/bulk_read_100_bytes.yaml]
*
[bulk_read_1_kbyte.yaml|https://github.com/stef1927/cstar_bulk_read_test/blob/master/bulk_read_1_kbyte.yaml]
*
[bulk_read_10_kbytes.yaml|https://github.com/stef1927/cstar_bulk_read_test/blob/master/bulk_read_10_kbytes.yaml]
Here are sample cassandra-stress commands to generate the data and read it back:
{code}
cassandra-stress user profile=bulk_read_1_kbyte.yaml ops\(insert=1\) n=5M -rate
threads=25
cassandra-stress user profile=bulk_read_1_kbyte.yaml
ops\(all_columns_tr_query=1\) -rate threads=25
{code}
was (Author: stefania):
In terms of establishing the canonical bulk reading workload, I suggest using
the profiles available in [this
benchmark|https://github.com/stef1927/cstar_bulk_read_test] with single
clustering and row sizes of approximately 100 bytes, 1 kb and 10 kb. These
correspond to the following user profiles:
*
[bulk_read_value50_cluster1.yaml|https://github.com/stef1927/cstar_bulk_read_test/blob/master/bulk_read_value50_cluster1.yaml]
*
[bulk_read_value500_cluster1.yaml|https://github.com/stef1927/cstar_bulk_read_test/blob/master/bulk_read_value500_cluster1.yaml]
*
[bulk_read_value5k_cluster1.yaml|https://github.com/stef1927/cstar_bulk_read_test/blob/master/bulk_read_value5k_cluster1.yaml]
Here are sample cassandra-stress commands to generate the data and read it back:
{code}
cassandra-stress user profile=bulk_read_value50_cluster1.yaml ops\(insert=1\)
n=5M -rate threads=25
cassandra-stress user profile=bulk_read_value50_cluster1.yaml
ops\(all_columns_tr_query=1\) -rate threads=25
{code}
> Establish and implement canonical bulk reading workload(s)
> ----------------------------------------------------------
>
> Key: CASSANDRA-10331
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10331
> Project: Cassandra
> Issue Type: Sub-task
> Reporter: Ariel Weisberg
> Assignee: Stefania
> Labels: docs-impacting
> Fix For: 3.0.5, 3.5
>
>
> Implement a client, use stress, or extend stress to a bulk reading workload
> that is indicative of the performance we are trying to improve.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
