sql server - Performance Improve on SQL Large table -
im having 260 columns table in sql server. when run "select count(*) table" taking 5-6 count. table contains close 90-100 million records 260 columns more 50 % column contains null. apart that, user can build dynamic sql query on table ui, searching 90-100 million records take time return results. there way improve find functionality on sql table filter criteria can , can any1 suggest me fastest way aggregate data on 25gb data .ui should hanged or timeout
investigate horizontal partitioning. query performance if can force users put partitioning key predicates.
try vertical partitioning, split 1 260-column table several tables fewer columns. put values commonly required 1 table. queries reference table(s) contain columns required. give more rows per page i.e. fewer pages per query.
you have high fraction of nulls. sparse columns may help, calculate percentages can hurt if inappropriate. there's so question on this.
filtered indexes , filtered statistics may useful if db runs similar queries.
Comments
Post a Comment