首頁 > 資料庫 > mysql教程 > Oracle 12.1.0.2新特性 Approximate Count Distinct

Oracle 12.1.0.2新特性 Approximate Count Distinct

WBOY
發布: 2016-06-07 16:39:14
原創
1272 人瀏覽過

[oracle@PD009 ~]$ grep -i approx_count oracle.strsettings for approx_count_distinct optimizationsqkaGBPushdown: estimated memory without GPD = groupSize (%.2f) * aclsum (%u) = %.2f; estimated memory with GPD = optHllEntry (%u) * # of appro

[oracle@PD009 ~]$ grep -i approx_count oracle.str
settings for approx_count_distinct optimizations
qkaGBPushdown:  estimated memory without GPD = groupSize (%.2f) * aclsum (%u) = %.2f; estimated memory with GPD = optHllEntry (%u) * # of approx_count_distinct (%d) * parallelDegree (%.0f) = %.2f
APPROX_COUNT_DISTINCT
APPROX_COUNT_DISTINCT
APPROX_COUNT_DISTINCT
_approx_cnt_distinct_gby_pushdown = choose
_approx_cnt_distinct_optimization = 0
 alter system flush shared_pool;
  alter session set events '10053 trace name context forever ,level 1';
 select count( distinct prod_id) from sales_history where  amount_sold>1;
 select approx_count_distinct(prod_id) from sales_history where  amount_sold>1;
  oradebug setmypid;
  oradebug tracefile_name;
  /s01/diag/rdbms/pdprod/PDPROD/trace/PDPROD_ora_4086.trc
  sql= select count( distinct prod_id) from sales_history where  amount_sold>1
----- Explain Plan Dump -----
----- Plan Table -----
============
Plan Table
============
---------------------------------------------+-----------------------------------+
| Id  | Operation             | Name         | Rows  | Bytes | Cost  | Time      |
---------------------------------------------+-----------------------------------+
| 0   | SELECT STATEMENT      |              |       |       |  4912 |           |
| 1   |  SORT AGGREGATE       |              |     1 |    13 |       |           |
| 2   |   VIEW                | VW_DAG_0     |    72 |   936 |  4912 |  00:00:59 |
| 3   |    HASH GROUP BY      |              |    72 |   648 |  4912 |  00:00:59 |
| 4   |     TABLE ACCESS FULL | SALES_HISTORY| 3589K |   32M |  4820 |  00:00:58 |
---------------------------------------------+-----------------------------------+
sql= select approx_count_distinct(prod_id) from sales_history where  amount_sold>1
----- Explain Plan Dump -----
----- Plan Table -----
============
Plan Table
============
----------------------------------------------+-----------------------------------+
| Id  | Operation              | Name         | Rows  | Bytes | Cost  | Time      |
----------------------------------------------+-----------------------------------+
| 0   | SELECT STATEMENT       |              |       |       |  4820 |           |
| 1   |  SORT AGGREGATE APPROX |              |     1 |     9 |       |           |
| 2   |   TABLE ACCESS FULL    | SALES_HISTORY| 3589K |   32M |  4820 |  00:00:58 |
----------------------------------------------+-----------------------------------+
Predicate Information: 
登入後複製

Related posts:

  1. Extract SQL Plan from AWR
  2. 部分行索引使用介绍
  3. Script:诊断解析等待和高version count
  4. 【11g新特性】SPM SQL PLAN MANAGEMENT执行计划管理流程图
  5. Number of distinct values (NDV) & synopsis & Histogram gather speed
  6. SQL调优:带函数的谓词导致CBO Cardinality计算误差
  7. 【11g新特性】Cardinality Feedback基数反馈
  8. 【Oracle CBO优化器】视图合并View Merging技术 _complex_view_merging & _simple_view_merging
  9. 海量数据插入性能测试
  10. 查询v$lock缓慢和direct path write temp等待
相關標籤:
來源:php.cn
本網站聲明
本文內容由網友自願投稿,版權歸原作者所有。本站不承擔相應的法律責任。如發現涉嫌抄襲或侵權的內容,請聯絡admin@php.cn
熱門教學
更多>
最新下載
更多>
網站特效
網站源碼
網站素材
前端模板