Kevin Yao created GRIFFIN-78:
--------------------------------
Summary: Manage job basic info by griffin rather than quartz
Key: GRIFFIN-78
URL: https://issues.apache.org/jira/browse/GRIFFIN-78
Project: Griffin (Incubating)
Issue Type: Improvement
Components: accuracy-batch
Affects Versions: 0.2.0-incubating
Reporter: Kevin Yao
In griffin now,jobs info is completely managed by quartz.Sometimes it is
inconvenient for us to get basic job info and delete job,because the structure
for quartz table is relatively complex.Therefore,We decide to create a simple
table in griffin to manage job basic info.
Our solution:
1.we create a base table called job.It only has four columns,which are*
id(primary ky)*,*job_name(varchar)*,*group_name(varchar)* and
*metric_name(varchar)*.
2.Different job type can inherit the base table.For instance,basic job
info can be stored in griffin_job table ,which inherits base table and has
other columns like measure_id(bigint),
quartz_job_name(varchar),quartz_group_name(varchar),deleted(bit).
3.This solution reduces redundant info and has a good expansibility.It
looks more reasonable,but it also increases complexity of table.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
