# VerticaBlog
**Repository Path**: null_334/VerticaBlog
## Basic Information
- **Project Name**: VerticaBlog
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No
## Statistics
- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-07-27
- **Last Updated**: 2020-12-19
## Categories & Tags
**Categories**: Uncategorized
**Tags**: None
## README
MyVertica 中文社区常见问题
MyVertica中文社区说明
==========
这里是Vertica爱好者关于Vertica相关技术交流、学习和互助的社区，不负责Vertica售后支持。
关于售后支持: 对于Vertica正式客户，建议中文邮件通过 china_vertica_support@hpe.com 咨询（verticahelp@hpe.com只接受英文邮件），他们是专职的中文支持团队，响应会更即时。
- china_vertica_support@hpe.com: 9x5, Chinese
- verticahelp@hpe.com: 7x24 for production issues, English
- production critical issue support: 1-877-669-1295 or 1-617-386-4500, English.
相关连接
----------
* [本文档(https://github.com/dingqiangliu/MyVertica/blob/master/README.md)](https://github.com/dingqiangliu/MyVertica/blob/master/README.md)
* [官方安装介质(https://my.vertica.com/download/vertica/)](https://my.vertica.com/download/vertica/)
* [官方产品文档(https://my.vertica.com/documentation/vertica/)](https://my.vertica.com/documentation/vertica/)
* [在线培训(英文字幕, https://my.vertica.com/resources/web-based-training/)](https://my.vertica.com/resources/web-based-training/)
* [动手训练(https://my.vertica.com/get-started-vertica/)](https://my.vertica.com/get-started-vertica/)
* [知识库(https://community.dev.hpe.com/t5/Vertica-Knowledge-Base/tkb-p/bigdata_wiki_vertica)](https://community.dev.hpe.com/t5/Vertica-Knowledge-Base/tkb-p/bigdata_wiki_vertica)
* [Marketplace(https://saas.hpe.com/marketplace/big-data#/Home/Show)](https://saas.hpe.com/marketplace/big-data#/Home/Show)
* [Vertica资源下载百度网盘(http://pan.baidu.com/s/1jXnMu#path=/vertica)](http://pan.baidu.com/s/1jXnMu#path=/vertica)
* [Vertica 专栏(http://vertica.club/)](http://vertica.club/)
常见问题: Vertica 常用问题 Checklists(更新中...)
==========
1. [节点复原(Node Recovery, https://my.vertica.com/node-recovery-checklist/)](https://my.vertica.com/node-recovery-checklist/)
2. [Spread调试(Spread Debugging, https://my.vertica.com/spread-debugging-checklist/)](https://my.vertica.com/spread-debugging-checklist/)
3. [数据库进程无法启动(Database Process Not Starting, https://my.vertica.com/database-process-not-starting/)](https://my.vertica.com/database-process-not-starting/)
5. [事务号管理(Epoch Management, https://my.vertica.com/checklist-draft/)](https://my.vertica.com/checklist-draft/)
6. [节点宕掉(Node Down, https://my.vertica.com/node-down-checklist/)](https://my.vertica.com/node-down-checklist/)
7. [Kafka(https://my.vertica.com/apache-kafka-not-ingesting-data/)](https://my.vertica.com/apache-kafka-not-ingesting-data/)
13. [停节点维护(Shutting down a node for maintenance, https://community.dev.hpe.com/t5/Vertica-Knowledge-Base/Prepare-HP-Vertica-Database-for-Maintenance/ta-p/229524)](https://community.dev.hpe.com/t5/Vertica-Knowledge-Base/Prepare-HP-Vertica-Database-for-Maintenance/ta-p/229524)
常见问题: 如何实现Oracle connectby相似功能？
==========
Connectby 实质上就是表自(外)关联和列转行。如果层级不多，是可以用标准SQL来改写。
其实用Vertica的SDK来实现相同功能的自定义函数，也不太难。参见：[Connectby for Veritca实现(http://pan.baidu.com/s/1jXnMu#path=/vertica/UDx/)](http://pan.baidu.com/s/1jXnMu#path=/vertica/UDx/) 下的 connectby-2013.08.30.tgz
对于Oracle中如下语句的执行效果:
select id, parent_id, name
, LEVEL as path_level
, CONNECT_BY_ROOT id AS id_root
, SYS_CONNECT_BY_PATH(id, '|') AS id_path
, CONNECT_BY_ROOT name AS name_root
, SYS_CONNECT_BY_PATH(name, '|') AS name_path
from company c
CONNECT BY PRIOR c.id = c.parent_id
START WITH c.parent_id = 0
order by id;
ID PARENT_ID NAME PATH_LEVEL ID_ROOT ID_PATH NAME_ROOT NAME_PATH
-----+-----------------+-----+----------+-----------+------------------------
1 0 Patrick1 1 1 |1 Patrick1 |Patrick1
2 0 Patrick2 1 2 |2 Patrick2 |Patrick2
12 1 Jim1 2 1 |1|12 Patrick1 |Patrick1|Jim1
13 1 Sandy1 2 1 |1|13 Patrick1 |Patrick1|Sandy1
14 13 Brian1 3 1 |1|13|14 Patrick1 |Patrick1|Sandy1|Brian1
15 13 Otto1 3 1 |1|13|15 Patrick1 |Patrick1|Sandy1|Otto1
22 2 Jim2 2 2 |2|22 Patrick2 |Patrick2|Jim2
23 2 Sandy2 2 2 |2|23 Patrick2 |Patrick2|Sandy2
24 23 Brian2 3 2 |2|23|24 Patrick2 |Patrick2|Sandy2|Brian2
25 23 Otto2 3 2 |2|23|25 Patrick2 |Patrick2|Sandy2|Otto2
10 rows selected.
select connectby(parent_id, id, name) over ()
from item;
level | parent_id | id | name | name_root | name_path
-------+-----------+-----+----------+-----------+------------------------
1 | 0 | 1 | Patrick1 | Patrick1 | Patrick1
1 | 0 | 2 | Patrick2 | Patrick2 | Patrick2
2 | 1 | 12 | Jim1 | Patrick1 | Patrick1/Jim1
2 | 1 | 13 | Sandy1 | Patrick1 | Patrick1/Sandy1
2 | 2 | 22 | Jim2 | Patrick2 | Patrick2/Jim2
2 | 2 | 23 | Sandy2 | Patrick2 | Patrick2/Sandy2
3 | 13 | 131 | Brian1 | Patrick1 | Patrick1/Sandy1/Brian1
3 | 13 | 132 | Otto1 | Patrick1 | Patrick1/Sandy1/Otto1
3 | 23 | 231 | Brian2 | Patrick2 | Patrick2/Sandy2/Brian2
3 | 23 | 232 | Otto2 | Patrick2 | Patrick2/Sandy2/Otto2
(10 rows)
create table if not exists fact
as /*+ direct */
select num as pk /* 主键 */, num%1000 as fk /* 构造各种外键或其他信息 */ from (
SELECT extract(epoch from ts)::int as num FROM (
SELECT '1970-01-01 00:00:01 +0'::TIMESTAMP AS tm
UNION
SELECT '1974-01-02 00:00:00 +0'::TIMESTAMP AS tm
) t0
TIMESERIES ts AS '1 second' OVER (ORDER BY tm)
) t1
order by pk
encoded by pk encoding DELTAVAL, fk encoding RLE
segmented by hash(pk) all nodes ksafe
;
select schema_name,anchor_table_name,p.projection_name,rowcount,start as start_epoch,"end" as end_epoch
from storage_containers sc, vs_wos_containers wc, projections p
where sc.storage_oid=wc.wosid and sc.projection_name=p.projection_name
order by start_epoch desc;