In addition to Weibo, there is also WeChat
Please pay attention
WeChat public account
Shulou
2025-02-23 Update From: SLTechnology News&Howtos shulou NAV: SLTechnology News&Howtos > Database >
Share
Shulou(Shulou.com)06/01 Report--
In the previous chapter, we discussed the segmentation of super-large tables through partitioning + parallelism, which can greatly improve the average distribution of data, but it is not the most perfect.
For example, to increase the amount of data by a few more levels, we assume that the table is currently 1T in size. There are 10 partitions, and the largest partition is 400 gigabytes, so if we want to export data as evenly as possible, using parallelism may not work as well.
For example, we require each dump file to be controlled at 200m, so 400G partitions need 800parallelism to complete. In actual database maintenance, we know that the default number of parallelism is only 64, which is several times higher than that of 800m.
Therefore, in the case of a large amount of data, if resources are tight, the dump generated may be relatively large.
We consider using rowid to meet our needs.
We can specify how many dump files we need to generate as needed. For example, if the table subscriber is 600m, then if we take 200m as a unit, we need to generate three dump files.
If you want the data to be average enough, you need to do some work on rowid.
Let's first set a parameter file in the following format.
You can see that the amount of data in the table memo is so large that according to 200m a unit, the largest partition (P9_A3000_E5) requires 800parallelism.
The table ICE_AGREEMENT is relatively small and is not a partition table. We use x as the synonym of the partition table, which can be easily identified when dealing with it.
MEMO P9_A3000_E0 156
MEMO P9_A3000_E1 170
MEMO P9_A3000_E2 190
MEMO P9_A3000_E3 200
MEMO P9_A3000_E4 180
MEMO P9_A3000_E5 800
MEMO PMAXVALUE_AMAXVALUE_EMAXVALUE 1
ICE_AGREEMENT x 36
CRIBER_HISTORY x 11
You can use the following script to complete the segmentation of the rowid.
# $1 dba conn details
# $2 table owner
# $3 table_name
# $4 subobject_name
# $5 parallel_no
Function normal_split
{
Sqlplus-s $1
Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.
Views: 0
*The comments in the above article only represent the author's personal views and do not represent the views and positions of this website. If you have more insights, please feel free to contribute and share.
Continue with the installation of the previous hadoop.First, install zookooper1. Decompress zookoope
"Every 5-10 years, there's a rare product, a really special, very unusual product that's the most un
© 2024 shulou.com SLNews company. All rights reserved.