January 08, 2016

PostgresSql Index Scripts


Today everything is about maintenace of  the postgres database so, searching and building scripts. Don't try to re-invent the wheel...
cheers


/****http://www.databasesoup.com/2014/05/new-finding-unused-indexes-query.html***************/

/**************unused index******************/
WITH table_scans as (
    SELECT relid,
        tables.idx_scan + tables.seq_scan as all_scans,
        ( tables.n_tup_ins + tables.n_tup_upd + tables.n_tup_del ) as writes,
                pg_relation_size(relid) as table_size
        FROM pg_stat_user_tables as tables
),
all_writes as (
    SELECT sum(writes) as total_writes
    FROM table_scans
),
indexes as (
    SELECT idx_stat.relid, idx_stat.indexrelid,
        idx_stat.schemaname, idx_stat.relname as tablename,
        idx_stat.indexrelname as indexname,
        idx_stat.idx_scan,
        pg_relation_size(idx_stat.indexrelid) as index_bytes,
        indexdef ~* 'USING btree' AS idx_is_btree
    FROM pg_stat_user_indexes as idx_stat
        JOIN pg_index
            USING (indexrelid)
        JOIN pg_indexes as indexes
            ON idx_stat.schemaname = indexes.schemaname
                AND idx_stat.relname = indexes.tablename
                AND idx_stat.indexrelname = indexes.indexname
    WHERE pg_index.indisunique = FALSE
),
index_ratios AS (
SELECT schemaname, tablename, indexname,
    idx_scan, all_scans,
    round(( CASE WHEN all_scans = 0 THEN 0.0::NUMERIC
        ELSE idx_scan::NUMERIC/all_scans * 100 END),2) as index_scan_pct,
    writes,
    round((CASE WHEN writes = 0 THEN idx_scan::NUMERIC ELSE idx_scan::NUMERIC/writes END),2)
        as scans_per_write,
    pg_size_pretty(index_bytes) as index_size,
    pg_size_pretty(table_size) as table_size,
    idx_is_btree, index_bytes
    FROM indexes
    JOIN table_scans
    USING (relid)
),
index_groups AS (
SELECT 'Never Used Indexes' as reason, *, 1 as grp
FROM index_ratios
WHERE
    idx_scan = 0
    and idx_is_btree
UNION ALL
SELECT 'Low Scans, High Writes' as reason, *, 2 as grp
FROM index_ratios
WHERE
    scans_per_write <= 1
    and index_scan_pct < 10
    and idx_scan > 0
    and writes > 100
    and idx_is_btree
UNION ALL
SELECT 'Seldom Used Large Indexes' as reason, *, 3 as grp
FROM index_ratios
WHERE
    index_scan_pct < 5
    and scans_per_write > 1
    and idx_scan > 0
    and idx_is_btree
    and index_bytes > 100000000
UNION ALL
SELECT 'High-Write Large Non-Btree' as reason, index_ratios.*, 4 as grp
FROM index_ratios, all_writes
WHERE
    ( case when total_writes = 0 then 0 else writes::NUMERIC / ( total_writes + 1 )end ) > 0.02
    AND NOT idx_is_btree
    AND index_bytes > 100000000
ORDER BY grp, index_bytes DESC )
SELECT reason, schemaname, tablename, indexname,
    index_scan_pct, scans_per_write, index_size, table_size
FROM index_groups;
/******************Table scan summary***********************************************/
SELECT
x1.table_in_trouble,
pg_relation_size(x1.table_in_trouble) AS sz_n_byts,
x1.seq_scan, x1.idx_scan,
CASE
WHEN pg_relation_size(x1.table_in_trouble) > 500000000
THEN 'Exceeds 500 megs, too large to count in a view. For a count, count individually'::text
ELSE count(x1.table_in_trouble)::text
END AS tbl_rec_count,
x1.priority
FROM
(
SELECT
(schemaname::text || '.'::text) || relname::text AS table_in_trouble,
seq_scan,
idx_scan,
CASE
WHEN (seq_scan - idx_scan) < 500 THEN 'Minor Problem'::text
WHEN (seq_scan - idx_scan) >= 500 AND (seq_scan - idx_scan) < 2500 THEN 'Major Problem'::text
WHEN (seq_scan - idx_scan) >= 2500 THEN 'Extreme Problem'::text
ELSE NULL::text
END AS priority
FROM
pg_stat_all_tables
WHERE
seq_scan > idx_scan
AND schemaname != 'pg_catalog'::name
AND seq_scan > 100) x1
GROUP BY
x1.table_in_trouble,
x1.seq_scan,
x1.idx_scan,
x1.priority
ORDER BY
x1.priority DESC,
x1.seq_scan
;
/********************Check Query Plan*********************/
explain ( analyze, BUFFERS)
select * from ?

/****Postgres index Summary Row Size************/
SELECT
    pg_class.relname,
    pg_size_pretty(pg_class.reltuples::bigint) AS rows_in_bytes,
    pg_class.reltuples AS num_rows,
    count(indexname) AS number_of_indexes,
    CASE WHEN x.is_unique = 1 THEN 'Y'
       ELSE 'N'
    END AS UNIQUE,
    SUM(case WHEN number_of_columns = 1 THEN 1
              ELSE 0
            END) AS single_column,
    SUM(case WHEN number_of_columns IS NULL THEN 0
             WHEN number_of_columns = 1 THEN 0
             ELSE 1
           END) AS multi_column
FROM pg_namespace
LEFT OUTER JOIN pg_class ON pg_namespace.oid = pg_class.relnamespace
LEFT OUTER JOIN
       (SELECT indrelid,
           max(CAST(indisunique AS integer)) AS is_unique
       FROM pg_index
       GROUP BY indrelid) x
       ON pg_class.oid = x.indrelid
LEFT OUTER JOIN
    ( SELECT c.relname AS ctablename, ipg.relname AS indexname, x.indnatts
           AS number_of_columns FROM pg_index x
           JOIN pg_class c ON c.oid = x.indrelid
           JOIN pg_class ipg ON ipg.oid = x.indexrelid  )
    AS foo
    ON pg_class.relname = foo.ctablename
WHERE
     pg_namespace.nspname='SchemaNameHere'
AND  pg_class.relkind = 'r'
GROUP BY pg_class.relname, pg_class.reltuples, x.is_unique
ORDER BY 2;













/****Postgres index Stats Summary******/
SELECT
    distinct
    t.tablename,
   indexname,
   c.reltuples AS num_rows,
   pg_size_pretty(pg_relation_size(t.schemaname::TEXT||'.'||quote_ident((t.tablename))::text)) AS table_size,
   pg_size_pretty(pg_relation_size(t.schemaname||'.'||quote_ident(indexrelname)::text)) AS index_size,
    CASE WHEN indisunique THEN 'Y'
       ELSE 'N'
    END AS UNIQUE,
    idx_scan AS number_of_scans,
    idx_tup_read AS tuples_read,
    idx_tup_fetch AS tuples_fetched
 FROM pg_tables t
LEFT OUTER JOIN pg_class c ON t.tablename=c.relname
LEFT OUTER JOIN
    ( SELECT c.relname AS ctablename, ipg.relname AS indexname, x.indnatts AS number_of_columns, idx_scan, idx_tup_read, idx_tup_fetch, indexrelname, indisunique FROM pg_index x
           JOIN pg_class c ON c.oid = x.indrelid
           JOIN pg_class ipg ON ipg.oid = x.indexrelid
           JOIN pg_stat_all_indexes psai ON x.indexrelid = psai.indexrelid )
    AS foo
    ON t.tablename = foo.ctablename
WHERE t.schemaname='SchemaNameHere'
ORDER BY 1,2;

No comments:

Post a Comment

Contact Form

Name

Email *

Message *