Uploaded bycookie1969
19 views

Hailey_Database_Performance_Made_Easy_through_Graphics.pdf

1. The document discusses using graphics and data visualization to improve understanding of database performance issues and SQL tuning. It provides examples of how visualizations can clearly show relationships in complex SQL queries and data that are difficult to understand from text or code alone. 2. Key steps in visual SQL tuning are laid out, including drawing tables as nodes, joins as connection lines, and filters as markings on tables. This helps identify optimization opportunities like missing indexes or stale statistics. 3. The document emphasizes that a lack of clarity in visualizing complex data and queries can have devastating consequences, while graphics enable easy understanding and effective problem-solving.

Embed presentation

1 DB Performance with Graphics With Kyle Hailey Product Manger for DB Optimizer OracleMonitor.com
2 Who is Kyle Hailey • 1990 Oracle – 90 support – 92 Ported v6 – 93 France – 95 Benchmarking – 98 ST Real World Performance • 2000 Dot.Com • 2001 Quest • 2002 Oracle OEM 10g • 2006 Independent • 2008 Embarcadero – DB Optimizer Success! First successful OEM design
3 Launch: Pressure Midnight before January 28, 1986 Lives are on the line Thanks to Edward Tufte
4 Copyright 2006 Kyle Hailey 13 Pages Faxed
5 Copyright 2006 Kyle Hailey Original Engineering data only showed damage “damages at the hottest and coldest temperature” - management
6 Copyright 2006 Kyle Hailey Congressional Hearings Evidence
7 Copyright 2006 Kyle Hailey Clearer 1. Include successes 2. Mark Differences 3. Normalize same temp 4. Scale known vs unknown 55 65 75 60 70 80 4 8 12 4 8 12 30 40 50 35 45 X
8 Copyright 2006 Kyle Hailey Difficult • NASA Engineers Fail • Congressional Investigators Fail • Data Visualization is Difficult But … Lack of Clarity can be devastating
9 Solutions • Clear Identification – Know how to identify problems and issues • Access to details – Provide solutions and/or information to address the issues • Graphics – Easy understanding, effective communication and discussion
10 First Step: Graphics “The humans … are exceptionally good at parsing visual information, especially when that information is coded by color and/or _____ .” Knowledge representation in cognitive science. Westbury, C. & Wilensky, U. (1998) motion
11 Why Use Graphics You can't imagine how many times I was told that nobody wanted or would use graphics … -- Jef Raskin, the creator of the Macintosh Infocus – (overhead projectors) sited a study that humans can parse graphical information 400,000 times faster than textual data
1212 10/30/10 Counties in US • 3101 Counties in US • 50 pages
13 “If I can't picture it, I can't understand it” Anscombe's Quartet I II III IV x y x y x y x y 10 8.04 10 9.14 10 7.46 8 6.58 8 6.95 8 8.14 8 6.77 8 5.76 13 7.58 13 8.74 13 12.74 8 7.71 9 8.81 9 8.77 9 7.11 8 8.84 11 8.33 11 9.26 11 7.81 8 8.47 14 9.96 14 8.1 14 8.84 8 7.04 6 7.24 6 6.13 6 6.08 8 5.25 4 4.26 4 3.1 4 5.39 19 12.5 12 10.84 12 9.13 12 8.15 8 5.56 7 4.82 7 7.26 7 6.42 8 7.91 5 5.68 5 4.74 5 5.73 8 6.89 Average 9 7.5 9 7.5 9 7.5 9 7.5 Standard Deviation 3.31 2.03 3.31 2.03 3.31 2.03 3.31 2.03 Linear Regression 1.33 1.33 1.33 1.33 - Albert Einstein
14 Graphics for Anscombe’s Quartet
15 What is a day in the life look like for a DBA who has performance issues? Tuning the Database Complex Averages Anscombe's Quartet I II III IV x y x y x y x y Average 9 7.5 9 7.5 9 7.5 9 7.5 Standard Deviation 3.31 2.03 3.31 2.03 3.31 2.03 3.31 2.03 Linear Regression 1.33 1.33 1.33 1.33
16 LOAD Top Activity SQL Events Sessions Get Details Click here Max CPU (yard stick) How Can We Open the Black Box?
17 How do we get our data? Database User 1 User 2 User 3 User 4 Active Sessions = 1 2 3 4 Graph represents # of sessions active, but also represents amount of time active in the database For Every Active Session there is a user (or application) waiting Sample v$session : fast and light weight We collect • session • SQL • State (CPU, IO, LOCK, Other Waits) • and many other columns Multi-dimensional fine grain data. Difficult to use w/o graphics
18 AWR collects 06/05/08 25 34 36 38 45 F1qcyh20550cf fj6gjgsshtxyx 0cjsxw5ndqdbc 8t8as9usk11qw dr1rkrznhh95b 10dkqv3kr8xa5 38zhkf4jdyff4 298wmz1kxjs1m CPU Enq: TX – row lock contention SQL*Net break/reset to client db file scattered read db file sequential read SQL (v$sqlstats) Sessions (v$sesstat) Waits (v$system_event) Which session executed which SQL? Which SQL was block on a wait event ? AWR ( and Statspack) Can’t answer! Old Method
19 After sampling Multi-dimensional 06/05/08 25 34 36 38 45 63 65 87 F1qcyh20550cf fj6gjgsshtxyx 0cjsxw5ndqdbc 8t8as9usk11qw dr1rkrznhh95b 10dkqv3kr8xa5 38zhkf4jdyff4 298wmz1kxjs1m CPU Enq: TX – row lock contention SQ L*N et break/reset to client db file scattered read db file sequential read IO Application SQL Sessions W aits Statistics Service Scott System User Program Sys Sqlplus SQL*Forms GL OE Package Procedure Plan Child # Module Action File, block, object p1, p2, p3 Sampling V$SESSION
20 DB Optimizer shows it all Copyright 2006 Kyle Hailey
21 21 10/30/10 Average Active Sessions (AAS) Use CPU count as yardstick:  AAS < 1 Database is not blocked  AAS ~= 0 Database basically idle Problems are in the APP not DB  AAS >> # of CPUS There is a bottleneck
22 Interpreting the Load Chart 1. Application – Code inefficient? 1. Database – Configured correctly? 1. Machine – Is the machine undersized? 1. SQL – Inefficient SQL? Chart answers this question and assists with solutions Who’s problem is it ?
23 1. Application Issues insert into foo values ('a'); commit; insert into foo values ('a'); commit; insert into foo values ('a'); commit; insert into foo values ('a'); commit; insert into foo values ('a'); commit; insert into foo values ('a'); commit; insert into foo values ('a'); commit; 4 concurrent sessions run :
24 2. Database issue begin for i in 1..1000 loop insert into foo values ('a'); end loop; end; / Commit; 4 concurrent sessions run :
25 3. Machine Undersized
26 4. SQL needs Tuning
27 Tuning SQL Complexity PARSING IN CURSOR #2 len=53 dep=0 uid=61 oct=3 lid=61 tim=1151519905950403 hv=2296704914 ad='4e50010c' SELECT 'Hello, world; today is ' || SYSDATE FROM dual END OF STMT PARSE #2:c=4000,e=1540,p=0,cr=0,cu=0,mis=1,r=0,dep=0,og=1, tim=1151519905950397 BINDS #2: EXEC #2:c=0,e=58,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=1,tim=1 151519906034782 WAIT #2: nam='SQL*Net message to client' ela= 2 driver id=1650815232 #bytes=1 p3=0 obj#=-1 tim=1151519906034809 FETCH #2:c=0,e=29,p=0,cr=0,cu=0,mis=0,r=1,dep=0,og=1,tim=1 151519906034864 WAIT #2: nam='SQL*Net message from client' ela= 215 driver id=1650815232 #bytes=1 p3=0 obj#=-1 tim=1151519906035133 FETCH #2:c=0,e=1,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=0,tim=11 51519906035165 WAIT #2: nam='SQL*Net message to client' ela= 1 driver id=1650815232 #bytes=1 p3=0 obj#=-1 tim=1151519906035188 WAIT #2: nam='SQL*Net message from client' ela= 192 driver id=1650815232 #bytes=1 p3=0 obj#=-1 tim=1151519906035400 STAT #2 id=1 cnt=1 pid=0 pos=1 obj=0 op='FAST DUAL (cr=0 pr=0 pw=0 time=3 us) -------------------------------------------------------------------------------------------- | Id | Operation | Name | Starts | E-Rows | A-Rows | -------------------------------------------------------------------------------------------- - | 1 | HASH GROUP BY | | 1 | 1 | 1 | |* 2 | FILTER | | 1 | | 1909 | |* 3 | TABLE ACCESS BY INDEX ROWID | PS_RETROPAYPGM_TBL | 1 | 1 | 3413 | | 4 | NESTED LOOPS | | 1 | 165 | 6827 | |* 5 | HASH JOIN | | 1 | 165 | 3413 | |* 6 | HASH JOIN | | 1 | 165 | 3624 | | 7 | TABLE ACCESS BY INDEX ROWID | WB_JOB | 1 | 242 | 2895 | | 8 | NESTED LOOPS | | 1 | 233 | 2897 | | 9 | TABLE ACCESS BY INDEX ROWID| PS_PAY_CALENDAR | 1 | 1 | 1 | |* 10 | INDEX RANGE SCAN | PS0PAY_CALENDAR | 1 | 1 | 1 | |* 11 | INDEX RANGE SCAN | WBBJOB_B | 1 | 286 | 2895 | |* 12 | TABLE ACCESS FULL | WB_RETROPAY_EARNS | 1 | 27456 | 122K| | 13 | TABLE ACCESS FULL | PS_RETROPAY_RQST | 1 | 13679 | 13679 | |* 14 | INDEX RANGE SCAN | PS#RETROPAYPGM_TBL | 3413 | 1 | 3413 | | 15 | SORT AGGREGATE | | 1791 | 1 | 1791 | | 16 | FIRST ROW | | 1791 | 1 | 1579 | |* 17 | INDEX RANGE SCAN (MIN/MAX) | WB_JOB_F | 1791 | 1 | 1579 | | 18 | SORT AGGREGATE | | 1539 | 1 | 1539 | | 19 | FIRST ROW | | 1539 | 1 | 1539 | |* 20 | INDEX RANGE SCAN (MIN/MAX) | WB_JOB_G | 1539 | 1 | 1539 Trace file Execution Plan Complexity !
28 Visual SQL Tuning
29 Tuning SQL • Indexes – Missing indexes • Table Stats – Stale statistics • Histograms – Recommendations • Execution Plan – DB Optimizer automatically finds better plans • Visual SQL Tuning diagrams
30 VST Steps 1. Tables – drawn as nodes 1. Joins – drawn as connector lines 1. Filters – mark on each table with filter in where clause 30 10/30/10
31 How to VST: Tables and Joins SELECT C.Phone_Number, C.Honorific, C.First_Name, C.Last_Name, C.Suffix, C.Address_ID, A.Address_ID, A.Street_Address_Line1, A.Street_Address_Line2, A.City_Name, A.State_Abbreviation, A.ZIP_Code, OD.Deferred_Shipment_Date, OD.Item_Count, ODT.Text, OT.Text, P.Product_Description, S.Shipment_Date FROM Orders O, Order_Details OD, Products P, Customers C, Shipments S, Addresses A, Code_Translations ODT, Code_Translations OT WHERE UPPER(C.Last_Name) LIKE :Last_Name||'%' AND UPPER(C.First_Name) LIKE :First_Name||'%' AND OD.Order_ID = O.Order_ID AND O.Customer_ID = C.Customer_ID AND OD.Product_ID = P.Product_ID(+) AND OD.Shipment_ID = S.Shipment_ID(+) AND S.Address_ID = A.Address_ID(+) AND O.Status_Code = OT.Code AND OT.Code_Type = 'ORDER_STATUS' AND OD.Status_Code = ODT.Code AND ODT.Code_Type = 'ORDER_DETAIL_STATUS' AND O.Order_Date > :Now - 366 ORDER BY C.Customer_ID, O.Order_ID DESC, S.Shipment_ID, OD.Order_Detail_ID; Copyright 2006 Kyle Hailey Tables Orders O, Order_Details OD, Products P, Customers C, Shipments S, Addresses A, Code_Translations ODT, Code_Translations OT Joins OD.Order_ID = O.Order_ID O.Customer_ID = C.Customer_ID OD.Product_ID = P.Product_ID(+) OD.Shipment_ID = S.Shipment_ID(+) S.Address_ID = A.Address_ID(+) O.Status_Code = OT.Code OD.Status_Code = ODT.Code Dan Tow – SQL TUNING Filters WHERE UPPER(C.Last_Name) LIKE :Last_Name||'%' AND UPPER(C.First_Name) LIKE :First_Name||'%' AND OT.Code_Type = 'ORDER_STATUS' AND O.Order_Date > :Now – 366 AND ODT.Code_Type = 'ORDER_DETAIL_STATUS'
32 Layout tables and connections Tables Orders O, Order_Details OD, Products P, Customers C, Shipments S, Addresses A, Code_Translations ODT, Code_Translations OT Joins OD.Order_ID = O.Order_ID O.Customer_ID = C.Customer_ID OD.Product_ID = P.Product_ID(+) OD.Shipment_ID = S.Shipment_ID(+) S.Address_ID = A.Address_ID(+) O.Status_Code = OT.Code OD.Status_Code = ODT.Code F S A OD P ODT C OT O
33 Unstructured Copyright 2006 Kyle Hailey Joins OD.Order_ID = O.Order_ID O.Customer_ID = C.Customer_ID OD.Product_ID = P.Product_ID(+) OD.Shipment_ID = S.Shipment_ID(+) S.Address_ID = A.Address_ID(+) O.Status_Code = OT.Code OD.Status_Code = ODT.Code Neater, but can you do anything with it? What’s the optimal execution path? S A OD P ODT C OT O
34 Parents and Children Copyright 2006 Kyle Hailey Joins OD.Order_ID = O.Order_ID O.Customer_ID = C.Customer_ID OD.Product_ID = P.Product_ID(+) OD.Shipment_ID = S.Shipment_ID(+) S.Address_ID = A.Address_ID(+) O.Status_Code = OT.Code OD.Status_Code = ODT.Code ODT OD P C A O S OT Primary Key (unique index) No index or non-unique Master Detail Structure the tree
35 VST – filters and best path Parent Child Parent Child Concept: 1. Start at most selective filter 2. Join down first, before joining upwards Filters WHERE UPPER(C.Last_Name) LIKE :Last_Name||'%' AND UPPER(C.First_Name) LIKE :First_Name||'%' AND OT.Code_Type = 'ORDER_STATUS‘ AND ODT.Code_Type = 'ORDER_DETAIL_STATUS' AND O.Order_Date > :Now – 366  Filters help determine best path F F 30% 0.02% 100% * (select count(*) from TAB where condition) --------------------------- (select count(*) from Tab) S A OD P ODT C OT O
36 F F 30% 0.02% S A OD P ODT C OT O VST – best path Note: Oracle only joins in one table to the previous result set F F
37 Cartesian SELECT A.BROKER_ID BROKER_ID, A.BROKER_LAST_NAME BROKER_LAST_NAME, A.BROKER_FIRST_NAME BROKER_FIRST_NAME, A.YEARS_WITH_FIRM YEARS_WITH_FIRM, C.OFFICE_NAME OFFICE_NAME, SUM (B.BROKER_COMMISSION) TOTAL_COMMISSIONS FROM BROKER A, CLIENT_TRANSACTION B, OFFICE_LOCATION C, INVESTMENT I WHERE A.BROKER_ID = B.BROKER_ID AND A.OFFICE_LOCATION_ID = C.OFFICE_LOCATION_ID GROUP BY A.BROKER_ID, A.BROKER_LAST_NAME, A.BROKER_FIRST_NAME, A.YEARS_WITH_FIRM, C.OFFICE_NAME;
38 Implied Cartesian select c.client_first_name, c.client_last_name, ct.action, ct.price, b.broker_last_name, b.broker_first_name, o.office_name from client_transaction ct, client c, broker b, office_location o where ct.price > 100 and b.broker_id=ct.broker_id and c.broker_id = b.broker_id and o.office_location_id = b.office_location_id
39 Diagram work for Many to One What about many to many?
40 Unstructured Copyright 2006 Kyle Hailey Joins OD.Order_ID = O.Order_ID O.Customer_ID = C.Customer_ID OD.Product_ID = P.Product_ID(+) OD.Shipment_ID = S.Shipment_ID(+) S.Address_ID = A.Address_ID(+) O.Status_Code = OT.Code OD.Status_Code = ODT.Code S A OD P ODT C OT O S A OD P ODT C OT O v v v v v v v v v v v v v v F F 30% 0.02%
41 Many-to-One vs Many-to-Many Predicate Filter 1 2 3 B -> C -> A go to A or C? Now what?
42 Adding Constraints SELECT COUNT (*) FROM b, c, a WHERE b.val2 = 100 AND a.val1 = b.id AND b.val1 = c.id; 58 logical reads alter table c add constraint c_pk unique (id); alter table b add constraint b_pk unique (id); 7 logical reads
43 Join sizes Join Sizes Use the Join sizes to determine path of least resistance
44 Look at 3 queries • Query 1 runs more than 24 hours • Query 2 outer joins and scalar subqueries • Query 3 create path not available to Oracle
45 Query 1 : Over 24 hours to run SELECT A0.zuchinis, A0.brocoli, C0.Oranges FROM ( SELECT A1.planted_date, A1.pears, A1.zuchinis, A1.brocoli FROM FOO.A A1, ( SELECT zuchinis, brocoli FROM FOO.A A2 WHERE pears = 'M' AND planted_date + 0 >= ADD_MONTHS ((SELECT MAX (planted_date) FROM FOO.B B1 WHERE pears = 'M' ), - 11) GROUP BY zuchinis, brocoli HAVING COUNT (*) = 12 ) i2 WHERE A1.planted_date = (SELECT MAX (planted_date) FROM FOO.B B2 WHERE pears = 'M' ) AND A1.pears = 'M' AND A1.zuchinis = i2.zuchinis (+) AND A1.brocoli = i2.brocoli (+) UNION SELECT A4.planted_date, A4.pears,A4.zuchinis,A4.brocoli FROM FOO.A A4 WHERE A4.planted_date >'01-OCT-08' and A4.planted_date <'03-OCT-08' AND A4.pears = 'D' AND A4.green_beans = '1' AND NOT EXISTS (SELECT * FROM FOO.A A5 WHERE pears = 'M' AND planted_date = (SELECT MAX (planted_date) FROM FOO.B B3 WHERE pears = 'M' ) AND A4.zuchinis = A5.zuchinis AND A4.brocoli = A5.brocoli) ) b, FOO.A A0, FOO.C C0, FOO.D D0, FOO.E E0 WHERE A0.planted_date >'01-OCT-08' and A0.planted_date <'03-OCT-08' AND A0.pears = 'D' AND A0.green_beans = '1' AND A0.zuchinis = b.zuchinis AND A0.brocoli = b.brocoli AND A0.planted_date = C0.planted_date AND A0.pears = C0.pears AND A0.zuchinis = C0.zuchinis AND A0.brocoli = C0.brocoli AND A0.planted_date = D0.planted_date AND A0.pears = D0.pears AND A0.harvest_size = D0.harvest_size AND C0.Oranges = D0.Oranges AND C0.apples = D0.apples AND (D0.lemons = 0 OR D0.lemons IS NULL) AND A0.planted_date = E0.planted_date AND A0.pears = E0.pears AND A0.harvest_size = E0.harvest_size AND C0.Oranges = E0.Oranges AND C0.apples = E0.apples AND (E0.lemons = 0 OR E0.lemons IS NULL) ORDER BY A0.zuchinis, A0.brocoli;
46 Visual SQL Diagram 9 secs
47 Default vs Tuned Default Tuned
48 Comparing Plans : 24 hours to 5 mins NL NL NL NL HJ HJ NL NL
49 Q2 49 10/30/10 SELECT CASE WHEN M.NYC IS NULL THEN (SELECT /*+ qb_name(qb1) */ MAX (Kona) FROM foo.F WHERE harvest_date = to_date('08/10/2008','dd/mm/yyyy') AND Argentina = TRIM ('D') AND Norway = F_OUTER.Norway ELSE M.NYC END AS NYC, CASE WHEN F_OUTER.Perth IS NULL THEN NULL ELSE (SELECT /*+ qb_name(qb2) */ Georgia FROM foo.P WHERE harvest_date = to_date('08/10/2008','dd/mm/yyyy') AND Argentina = TRIM ('D') AND Paris = F_OUTER.Perth) END AS richard, CASE WHEN F_OUTER.Aruba IS NULL THEN NULL ELSE (SELECT /*+ qb_name(qb3) */ Georgia FROM foo.P WHERE harvest_date = to_date('08/10/2008','dd/mm/yyyy') AND Argentina = TRIM ('D') AND Paris = F_OUTER.Aruba) END AS Jody, CASE WHEN F_OUTER.Portland IS NULL THEN NULL ELSE (SELECT /*+ qb_name(qb4) */ Georgia FROM foo.P WHERE harvest_date = to_date('08/10/2008','dd/mm/yyyy') AND Argentina = TRIM ('D') AND Paris = F_OUTER.Portland) END AS Tom FROM foo.F F_OUTER, foo.M , foo.J , foo.N , (SELECT /*+ qb_name(qb5) */ H.SF, Oregon, H.Haiti, K.Bermuda, L.Denmark FROM (foo.H LEFT OUTER JOIN foo.K ON H.harvest_date = K.harvest_date AND H.Argentina = K.Argentina AND H.SF = K.SF AND K.Dallas = '001') LEFT OUTER JOIN FOo.L ON H.harvest_date = L.harvest_date AND H.Argentina = L.Argentina AND H.SF = L.SF WHERE H.harvest_date = to_date('08/10/2008','dd/mm/yyyy') AND H.Argentina = TRIM ('D')) extra WHERE F_OUTER.harvest_date = M.harvest_date(+) AND F_OUTER.Argentina = M.Argentina(+) AND F_OUTER.Norway = M.Norway(+) AND M.Norway(+) = M.Texas(+) AND F_OUTER.harvest_date = to_date('08/10/2008','dd/mm/yyyy') AND F_OUTER.Argentina = TRIM ('D') AND M.harvest_date(+) = to_date('08/10/2008','dd/mm/yyyy') AND M.Argentina(+) = TRIM ('D') AND F_OUTER.Norway = F_OUTER.Hawaii AND F_OUTER.harvest_date = J.harvest_date(+) AND F_OUTER.Argentina = J.Argentina(+) AND F_OUTER.Norway = J.Texas(+) AND J.harvest_date(+) = to_date('08/10/2008','dd/mm/yyyy') AND J.Argentina(+) = TRIM ('D') AND F_OUTER.Iraq = extra.SF(+) AND F_OUTER.harvest_date = N.harvest_date(+) Two important qualities: • All outer joins to F_OUTER • 4 subqueries in select select
50 Q2 10/30/10 12 825 845 682348 select
51 Q2 51 10/30/10 The subqueries in the select clause look like select CASE WHEN F.f1 IS NULL THEN NULL ELSE (SELECT X.f2 FROM X WHERE code_vl = F.f1) END AS f0 from F; and should be merged into the query like: select CASE WHEN F.f1 IS NULL THEN NULL ELSE ( X.f2) END AS f0 from F , X where code_vl(+) = F.f1; select CASE WHEN F.f1 IS NULL THEN NULL ELSE (SELECT X.f2 FROM X WHERE code_vl = F.f2) END AS f0 from F; select CASE WHEN F.f1 IS NULL THEN NULL ELSE ( X.f2) END AS f0 from F , X where code_vl(+) = F.f1;
52 Q3 52 10/30/10 SELECT DISTINCT * FROM FOO.a a, FOO.c c, FOO.d d, FOO.g g WHERE a.planted_date > '01-OCT-08' AND a.planted_date < '03-OCT-08' AND a.pears = 'D' AND a.green_beans = '1' AND a.planted_date = c.planted_date AND a.pears = c.pears AND a.zuchinis = c.zuchinis AND a.brocoli = c.brocoli AND a.planted_date = d.planted_date AND a.pears = d.pears AND a.harvest_size = d.harvest_size AND c.oranges = d.oranges AND c.apples = d.apples AND (d.lemons = 0 OR d.lemons IS NULL) AND a.planted_date = g.planted_date AND a.pears = g.pears AND a.harvest_size = g.harvest_size AND c.oranges = g.oranges AND c.apples = g.apples AND (g.lemons = 0 OR g.lemons IS NULL) AND a.zuchinis = '0236' AND d.apples = g.apples AND d.oranges = g.oranges ORDER BY a.zuchinis, a.brocoli;
53 Q3: Transitivity 53 10/30/10 1,126,402 7,136,362
54 Q3 54 10/30/10 SELECT * FROM ( SELECT /*+ NO_MERGE */ c.apples, c.oranges, a.harvest_size FROM a, c WHERE a.planted_date = TO_DATE ('02/10/2008', 'dd/mm/yyyy') AND a.pears = 'D' AND a.green_beans = '1' AND a.planted_date = c.planted_date AND a.pears = c.pears AND a.zuchinis = c.zuchinis AND a.brocoli = c.brocoli AND a.zuchinis = '0236' ) X, ( SELECT /*+ NO_MERGE */ d.apples, d.oranges, d.harvest_size FROM d, g WHERE d.planted_date = TO_DATE ('02/10/2008', 'dd/mm/yyyy') AND g.planted_date = TO_DATE ('02/10/2008', 'dd/mm/yyyy') AND g.apples = d.apples AND d.oranges = g.oranges AND d.pears = 'D' AND g.pears = 'D' AND g.pears = d.pears AND g.harvest_size = d.harvest_size AND (d.lemons = 0 OR d.lemons IS NULL) AND (g.lemons = 0 OR g.lemons IS NULL) ) Y WHERE X.oranges = Y.oranges AND X.apples = Y.apples AND X.harvest_size = Y.harvest_size; This final version runs in elapsed 0.33 secs and 12K logical reads down from an original elapsed 4.5 secs and 1M logical reads
55 What tools to use to create VST diagrams? • Paper – Modifications messy and difficult • PowerPoint – Can move things around – Sticky connectors – Easy to modify • DB Optimizer – Automatic and still modifiable
56 Visual SQL Tuning (VST) Table Sizes Join Sizes Filter Ratios Exists
57 VST Steps Summary 1.Diagram tables 2.Draw connectors for each join 3.Calculate filter ratios 4.Find the table sizes 5.Calculate two table join sizes Execution Path • Start at the most selective join filter • Join to keep the running result set size small Many to Many relationships = problems 57 10/30/10 Parent Child Parent Child F F 30% 0.02% S A OD P ODT C OT O
58 Copyright 2006 Kyle Hailey Do You Want? Engineering Data? Engineering Data?
59 Copyright 2006 Kyle Hailey Do You Want? Pretty Pictures Pretty Pictures
60 Copyright 2006 Kyle Hailey Do You Want? Clean and Clear Clean and Clear ? ? ? ? ? ?
61 Copyright 2006 Kyle Hailey Imagine Trying to Drive your Car And is updated once and hour And is updated once and hour Or would you like it to look … Would you want your dashboard to look like :
62 Copyright 2006 Kyle Hailey Or This
63 Summary 1.Database - AAS – Profile database – Use wait interface and graphics – Identify machine, application, database or SQL 1.SQL - VST – Indexes, stats, execution path – Visual SQL Tuning
64 Bibliography http://www.simple-talk.com/sql/performance/designing-efficient-sql-a-visual-approach/ Refactoring SQL Applications – Stephane Faroult Troubleshooting Oracle Performance – Christian Antognini SQL Tuning – Dan Tow Cost-Based Oracle Fundamentals – Jonathan Lewis
65 Visual SQL Tuning (VST) Industry Exclusive
66 DB Optimizer XE Key Features Profile Tune Load Test SQL IDE
67 DB Optimizer XE Key Features Profile Tune Load Test SQL IDE
68 DB Optimizer XE Key Features Profile Tune Load Test SQL IDE
69 DB Optimizer XE Key Features Profile Tune Load Test SQL IDE
70 Review of Functionality Profile Tune Load Test SQL IDE Only tool on the market with these features integrated
71 Thank You Kyle Hailey http://oraclemonitor.com
72 Appendix • Diagramming – Simple joins and inline views – Outer joins – Exists/not exists (in/not in) – Correlated aggregate sub-queries • Hints – LEADING – USE_NL – USE_HASH – INDEX – NO_MERGE • Execution Order • Machine Health • VST method with Statistics
73 Basic Joins Many to single value Cartesian Cartesian Safe Dangerous M
74 Simple queries and sub-queries 74 A B select * from A, B where b.f1=a.f1 Non-correlated sub-query select * from A,(select b.f1 from B) c where c.f1=a.f1 Simple join select * from A where a.f1 = (select max(b.f1) from B ) Special case: non correlated sub-query returns one row A B A B A B Unique index on b.f1 No unique indexes Unique index on Both b.f1 and a.f1 M
75 Outer Joins 75 10/30/10 Outer Joins
76 Correlated scalar sub-queries 76 10/30/10 Correlated aggregate subquery: select ename from emp a where a.sal > (select avg(sal) from emp b where a.deptno=b.deptno) A B Scalar Subqueries select ename, (select avg(sal) from emp b where a.deptno=b.deptno) from emp a;
77 Exists and Not In 77 10/30/10 SELECT d.* FROM dept d WHERE exists ( SELECT null FROM emp e WHERE e.deptno=d.deptno); SELECT d.* FROM dept d WHERE d.deptno in ( SELECT deptno FROM emp e ); select distinct d.* from dept1 d ,emp e where e.deptno = d.deptno; SELECT d.* FROM dept d WHERE not exists ( SELECT null FROM emp e WHERE e.deptno=d.deptno); SELECT d.* FROM dept d WHERE d.deptno not in ( SELECT deptno FROM emp e where e.deptno is not null ) or d.deptno is null; select d.* from dept1 d left outer join emp e on e.deptno = d.deptno where e.deptno is null; E N
78 HINTS • ORDERED - good on 9i • Leading(tab_alias , table_alias … ) – 10g format • USE_NL(table_alias) – Inner Table (not driving) • USE_HASH(table_alias) – 2cd table, probe into • INDEX(tab_alias index_name) • NO_MERGE Oracle first decides join order then join type (example http://www.adp-gmbh.ch/blog/2008/01/17.php) 78 10/30/10
79 Visual SQL Tuning (VST) diagrams 79 10/30/10 HJ NL i io HJ NL Makes comparison of execution plans easy Better yet, it will help us find the best execution path
80 SELECT O.ORDER_ID, LINE_ITEM_ID, PRODUCT_ID, UNIT_PRICE, QUANTITY, ORDER_MODE, ORDER_STATUS, ORDER_TOTAL, SALES_REP_ID, PROMOTION_ID, C.CUSTOMER_ID, CUST_FIRST_NAME, CUST_LAST_NAME, CREDIT_LIMIT, CUST_EMAIL, ORDER_DATE FROM ORDERS O, ORDER_ITEMS OI, CUSTOMERS C WHERE O.ORDER_ID = OI.ORDER_ID AND O.CUSTOMER_ID = C.CUSTOMER_ID AND O.ORDER_STATUS <= 4 VST vs Explain Plan 80 10/30/10 i MJ HJ HJ NL
81 i MJ HJ Bad Plans vs Good Plan 81 10/30/10 HJ NL
82 Comparing Plans : 24 hours to 5 mins NL NL NL NL HJ HJ NL NL
83 When to Tune 1. Machine a) CPU • Response times skewed • 100% CPU might be fine • Users wait in queue (run queue) => machine underpowered a) Memory • Paging • Wait times skewed (ex : latch free) • Erratic response times ( ex : ls ) 1. Oracle 1) Waits > CPU ? • tune waits 1) CPU > 100% ? • tune top CPU SQL 1) Else • It’s the application Oracle Load (AAS) Top Session Top Wait Top SQL SQL Detail Session Detail File Detail Object Detail Wait Detail Host AAS > #CPU AAS > 1 Waits > CPU CPU > Waits CPU Memory
84 Machine Make sure the machine is healthy before tuning Oracle • CPU => use run queue, < 2 * #CPU • Memory => page out VMSTAT
85 Summary 1.Machine - vmstat – Memory, CPU (we can see IO response in Oracle) 1.Database - AAS – Use wait interface and graphics – Identify machine, application, database or SQL 1.SQL - VST – Indexes, stats, execution path – Visual SQL Tuning
86 Solution to Many to Many Table Sizes Filter Ratios Join Sizes
87 VST Steps Objects: 1. Tables : drawn as nodes 2. Joins : drawn as connector lines 3. Filters : mark on each table with filter in where clause Statistics: 1.Table sizes 2. Join sizes 3. Calculate filter percentages – filter ratio = number of rows returned with filter / number of rows 87 10/30/10
88 Google Directions • Compare two sets of directions • (optionally show graphical set of directions) • Show map • Show map with traffic flow, read, yellow, green
89 How Can We Open the Black Box? OEM ASHMON/SASH DB Optimizer •Powerful - Identifies issues quickly and powerfully •Interactive - Allows exploring the data •Easy - Understandable by everyone, DBA, Dev and Managers !
90 Copyright 2006 Kyle Hailey Clearer 1. Include successes 2. Remove Irrelevant 3. Normalize same temp 4. Scale known vs unknown 55 65 75 60 70 80 4 8 12 4 8 12 30 40 50 35 45 X

More Related Content

PDF
31063115_1679409488310Developer_Tuning_Tips_-_UTOUG_Mar_2023.pdf
byTricantinoLopezPerez
 
PPTX
Top 10 tips for Oracle performance
byGuy Harrison
 
PDF
Chasing the optimizer
byMauro Pagano
 
PDF
Database and application performance vivek sharma
byaioughydchapter
 
PPTX
Oracle Query Optimizer - An Introduction
byadryanbub
 
PPTX
Oracle database performance tuning
byYogiji Creations
 
PPTX
Sql and PL/SQL Best Practices I
byCarlos Oliveira
 
PDF
Managing Statistics for Optimal Query Performance
byKaren Morton
 
31063115_1679409488310Developer_Tuning_Tips_-_UTOUG_Mar_2023.pdf
byTricantinoLopezPerez
 
Top 10 tips for Oracle performance
byGuy Harrison
 
Chasing the optimizer
byMauro Pagano
 
Database and application performance vivek sharma
byaioughydchapter
 
Oracle Query Optimizer - An Introduction
byadryanbub
 
Oracle database performance tuning
byYogiji Creations
 
Sql and PL/SQL Best Practices I
byCarlos Oliveira
 
Managing Statistics for Optimal Query Performance
byKaren Morton
 

Similar to Hailey_Database_Performance_Made_Easy_through_Graphics.pdf

PPTX
Oracle Database Performance Tuning Basics
bynitin anjankar
 
PPTX
Understanding Query Optimization with ‘regular’ and ‘Exadata’ Oracle
byGuatemala User Group
 
PDF
NOCOUG_201311_Fine_Tuning_Execution_Plans.pdf
bycookie1969
 
PDF
Advanced tips for making Oracle databases faster
bySolarWinds
 
PPTX
Presentación Oracle Database Migración consideraciones 10g/11g/12c
byRonald Francisco Vargas Quesada
 
PPT
Database performance tuning and query optimization
byUsman Tariq
 
PDF
dd presentation.pdf
byAnSHiKa187943
 
PPT
Cost Based Optimizer - Part 1 of 2
byMahesh Vallampati
 
PPTX
Qure Tech Presentation
bysqlserver.co.il
 
PPTX
Oracle performance tuning_sfsf
byMao Geng
 
PPTX
Beginners guide to_optimizer
byMaria Colgan
 
PPT
Oracle 10g Performance: chapter 00 intro live_short
byKyle Hailey
 
PPT
SQL Optimization With Trace Data And Dbms Xplan V6
byMahesh Vallampati
 
PPTX
DB
bySamchu Li
 
PDF
Brad McGehee Intepreting Execution Plans Mar09
byguest9d79e073
 
PDF
Brad McGehee Intepreting Execution Plans Mar09
byMark Ginnebaugh
 
PPTX
My Query is slow, now what?
byGianluca Sartori
 
PPTX
DOAG: Visual SQL Tuning
byKyle Hailey
 
PDF
Query Tuning for Database Pros & Developers
byCode Mastery
 
PPTX
Oracle performance tuning for java developers
bySaeed Shahsavan
 
Oracle Database Performance Tuning Basics
bynitin anjankar
 
Understanding Query Optimization with ‘regular’ and ‘Exadata’ Oracle
byGuatemala User Group
 
NOCOUG_201311_Fine_Tuning_Execution_Plans.pdf
bycookie1969
 
Advanced tips for making Oracle databases faster
bySolarWinds
 
Presentación Oracle Database Migración consideraciones 10g/11g/12c
byRonald Francisco Vargas Quesada
 
Database performance tuning and query optimization
byUsman Tariq
 
dd presentation.pdf
byAnSHiKa187943
 
Cost Based Optimizer - Part 1 of 2
byMahesh Vallampati
 
Qure Tech Presentation
bysqlserver.co.il
 
Oracle performance tuning_sfsf
byMao Geng
 
Beginners guide to_optimizer
byMaria Colgan
 
Oracle 10g Performance: chapter 00 intro live_short
byKyle Hailey
 
SQL Optimization With Trace Data And Dbms Xplan V6
byMahesh Vallampati
 
DB
bySamchu Li
 
Brad McGehee Intepreting Execution Plans Mar09
byguest9d79e073
 
Brad McGehee Intepreting Execution Plans Mar09
byMark Ginnebaugh
 
My Query is slow, now what?
byGianluca Sartori
 
DOAG: Visual SQL Tuning
byKyle Hailey
 
Query Tuning for Database Pros & Developers
byCode Mastery
 
Oracle performance tuning for java developers
bySaeed Shahsavan
 

More from cookie1969

PPT
active_session_history_oracle_performance.ppt
bycookie1969
 
PPT
copenhagen_schema_performance_tuning.ppt
bycookie1969
 
PDF
On Seeing Double in V$SQL_Thomas_Kytepdf
bycookie1969
 
PDF
Donatone_Grid Performance(2)11111111.pdg
bycookie1969
 
PPT
dokumen.tips_oracle-10g-advanced-performance-tuning-kyle-hailey-kylelfgmailco...
bycookie1969
 
PDF
200603ash.pdf Performance Tuning Oracle DB
bycookie1969
 
PDF
kharchenko-outliers-140211120833-phpapp01.pdf
bycookie1969
 
PPT
WLCG_Oracle_perf_for_admin_Luca_Nov07.ppt
bycookie1969
 
PDF
Maris+Elsins+-+Mining+AWR+V2(1)11111.pdg
bycookie1969
 
PDF
Identify heaviest plan_line_id for a given execution plan (Oracl.pdf
bycookie1969
 
PDF
ohio11_grid_kramer_present.pdf Statspack
bycookie1969
 
PDF
504_senegacnik_oracle_database_mutexes.pdf
bycookie1969
 
PDF
pdf-download-db-time-based-oracle-performance-tuning-theory-and.pdf
bycookie1969
 
PDF
What is new in the Oracle 9i CBOptimizer.pdf
bycookie1969
 
PDF
O_Need-for-Speed_Top-Five-Oracle-Performance-Tuning-Tips_NYOUG.pdf
bycookie1969
 
PPTX
2014_599_Rice_ppt_Weblogic_Oracle_Server.pptx
bycookie1969
 
PPT
ash_feel_the_power_kyle_hailey_8951122.ppt
bycookie1969
 
PPT
waits_events_and_the_geeks6043265(3).ppt
bycookie1969
 
PDF
con9577-ash-deep-dive-oow2013-2031468.pdf
bycookie1969
 
PDF
What is new in the Oracle 9i Optimizer.pdf
bycookie1969
 
active_session_history_oracle_performance.ppt
bycookie1969
 
copenhagen_schema_performance_tuning.ppt
bycookie1969
 
On Seeing Double in V$SQL_Thomas_Kytepdf
bycookie1969
 
Donatone_Grid Performance(2)11111111.pdg
bycookie1969
 
dokumen.tips_oracle-10g-advanced-performance-tuning-kyle-hailey-kylelfgmailco...
bycookie1969
 
200603ash.pdf Performance Tuning Oracle DB
bycookie1969
 
kharchenko-outliers-140211120833-phpapp01.pdf
bycookie1969
 
WLCG_Oracle_perf_for_admin_Luca_Nov07.ppt
bycookie1969
 
Maris+Elsins+-+Mining+AWR+V2(1)11111.pdg
bycookie1969
 
Identify heaviest plan_line_id for a given execution plan (Oracl.pdf
bycookie1969
 
ohio11_grid_kramer_present.pdf Statspack
bycookie1969
 
504_senegacnik_oracle_database_mutexes.pdf
bycookie1969
 
pdf-download-db-time-based-oracle-performance-tuning-theory-and.pdf
bycookie1969
 
What is new in the Oracle 9i CBOptimizer.pdf
bycookie1969
 
O_Need-for-Speed_Top-Five-Oracle-Performance-Tuning-Tips_NYOUG.pdf
bycookie1969
 
2014_599_Rice_ppt_Weblogic_Oracle_Server.pptx
bycookie1969
 
ash_feel_the_power_kyle_hailey_8951122.ppt
bycookie1969
 
waits_events_and_the_geeks6043265(3).ppt
bycookie1969
 
con9577-ash-deep-dive-oow2013-2031468.pdf
bycookie1969
 
What is new in the Oracle 9i Optimizer.pdf
bycookie1969
 

Recently uploaded

PPTX
Joseph Hill Honors Thesis Presentation 2024
byJoeHill71
 
PPT
Verilog - FPGA data path controller Design Sequential circuit FSM based
bykirtipfraoa
 
PDF
Module-3_Deep Learning_22ISE74A_ISE_.pdf
byDr. Shivashankar
 
PPTX
Introduction to Relational Algebra Advance Database Part three.pptx
bywilliamsmareh
 
PDF
Workshop practice theory course (Unit-1) By Varun Pratap Singh.pdf
byVarun Pratap Singh
 
PPTX
Unveiling the Myths & Truths: AI’s Impact in Engineering
byLuca Bonmassar
 
PPTX
Introduction to IoT Unit DATA ANALYTICS AND SUPPORTING SERVICES- IV.pptx
bygirishgandhi4
 
PPTX
33 KV SUBSTATION.pptx FROM LAYOUT TO PROTECTION
byPradipChanda5
 
PPTX
Planning and selection of equipment for different mining conditions. Equipmen...
bygugulothuakhil728
 
PPTX
thermal_design_project_work_internship(updated)-1.pptx
byshahveer210504
 
PPTX
Modularity In Lyra - A new studio's experiences building their prototype on U...
byiamscottcstevens
 
DOCX
Kabirpawarpptof java practicalexp1 to 11.docx
bynkpawar4413
 
PDF
Event #3_ Build a Gemini Bot, Together with GitHub_private.pdf
by3526ming
 
PPTX
ATD.pptx traction distribution railway atd
bykhuranarahul599
 
PDF
HYDRAULIC TURBINES with question and answer.pdf
bysparunkumar1
 
PPTX
Supercapacitor.pptx...............................
byTarikBouchala
 
PDF
MODULE FOUR THEODOLITE SURVEY SY CIVIL 5449
byDr. BASWESHWAR JIRWANKAR
 
PPT
6.1POWER BLOCK NEW.ppt indian railway three
bykhuranarahul599
 
PPTX
TRAY DRYER by Gulshan Athbhaiya.pptx
byGulshan Athbhaiya
 
PPTX
William Mulholland and the St. Francis Dam Disaster
byBob Mayer
 
Joseph Hill Honors Thesis Presentation 2024
byJoeHill71
 
Verilog - FPGA data path controller Design Sequential circuit FSM based
bykirtipfraoa
 
Module-3_Deep Learning_22ISE74A_ISE_.pdf
byDr. Shivashankar
 
Introduction to Relational Algebra Advance Database Part three.pptx
bywilliamsmareh
 
Workshop practice theory course (Unit-1) By Varun Pratap Singh.pdf
byVarun Pratap Singh
 
Unveiling the Myths & Truths: AI’s Impact in Engineering
byLuca Bonmassar
 
Introduction to IoT Unit DATA ANALYTICS AND SUPPORTING SERVICES- IV.pptx
bygirishgandhi4
 
33 KV SUBSTATION.pptx FROM LAYOUT TO PROTECTION
byPradipChanda5
 
Planning and selection of equipment for different mining conditions. Equipmen...
bygugulothuakhil728
 
thermal_design_project_work_internship(updated)-1.pptx
byshahveer210504
 
Modularity In Lyra - A new studio's experiences building their prototype on U...
byiamscottcstevens
 
Kabirpawarpptof java practicalexp1 to 11.docx
bynkpawar4413
 
Event #3_ Build a Gemini Bot, Together with GitHub_private.pdf
by3526ming
 
ATD.pptx traction distribution railway atd
bykhuranarahul599
 
HYDRAULIC TURBINES with question and answer.pdf
bysparunkumar1
 
Supercapacitor.pptx...............................
byTarikBouchala
 
MODULE FOUR THEODOLITE SURVEY SY CIVIL 5449
byDr. BASWESHWAR JIRWANKAR
 
6.1POWER BLOCK NEW.ppt indian railway three
bykhuranarahul599
 
TRAY DRYER by Gulshan Athbhaiya.pptx
byGulshan Athbhaiya
 
William Mulholland and the St. Francis Dam Disaster
byBob Mayer
 

Hailey_Database_Performance_Made_Easy_through_Graphics.pdf

  • 1.
    1 DB Performance with Graphics With KyleHailey Product Manger for DB Optimizer OracleMonitor.com
  • 2.
    2 Who is KyleHailey • 1990 Oracle – 90 support – 92 Ported v6 – 93 France – 95 Benchmarking – 98 ST Real World Performance • 2000 Dot.Com • 2001 Quest • 2002 Oracle OEM 10g • 2006 Independent • 2008 Embarcadero – DB Optimizer Success! First successful OEM design
  • 3.
    3 Launch: Pressure Midnight before January28, 1986 Lives are on the line Thanks to Edward Tufte
  • 4.
  • 5.
    5 Copyright 2006 Kyle Hailey OriginalEngineering data only showed damage “damages at the hottest and coldest temperature” - management
  • 6.
  • 7.
    7 Copyright 2006 Kyle Hailey Clearer 1.Include successes 2. Mark Differences 3. Normalize same temp 4. Scale known vs unknown 55 65 75 60 70 80 4 8 12 4 8 12 30 40 50 35 45 X
  • 8.
    8 Copyright 2006 Kyle Hailey Difficult •NASA Engineers Fail • Congressional Investigators Fail • Data Visualization is Difficult But … Lack of Clarity can be devastating
  • 9.
    9 Solutions • Clear Identification –Know how to identify problems and issues • Access to details – Provide solutions and/or information to address the issues • Graphics – Easy understanding, effective communication and discussion
  • 10.
    10 First Step: Graphics “Thehumans … are exceptionally good at parsing visual information, especially when that information is coded by color and/or _____ .” Knowledge representation in cognitive science. Westbury, C. & Wilensky, U. (1998) motion
  • 11.
    11 Why Use Graphics Youcan't imagine how many times I was told that nobody wanted or would use graphics … -- Jef Raskin, the creator of the Macintosh Infocus – (overhead projectors) sited a study that humans can parse graphical information 400,000 times faster than textual data
  • 12.
    1212 10/30/10 Counties inUS • 3101 Counties in US • 50 pages
  • 13.
    13 “If I can'tpicture it, I can't understand it” Anscombe's Quartet I II III IV x y x y x y x y 10 8.04 10 9.14 10 7.46 8 6.58 8 6.95 8 8.14 8 6.77 8 5.76 13 7.58 13 8.74 13 12.74 8 7.71 9 8.81 9 8.77 9 7.11 8 8.84 11 8.33 11 9.26 11 7.81 8 8.47 14 9.96 14 8.1 14 8.84 8 7.04 6 7.24 6 6.13 6 6.08 8 5.25 4 4.26 4 3.1 4 5.39 19 12.5 12 10.84 12 9.13 12 8.15 8 5.56 7 4.82 7 7.26 7 6.42 8 7.91 5 5.68 5 4.74 5 5.73 8 6.89 Average 9 7.5 9 7.5 9 7.5 9 7.5 Standard Deviation 3.31 2.03 3.31 2.03 3.31 2.03 3.31 2.03 Linear Regression 1.33 1.33 1.33 1.33 - Albert Einstein
  • 14.
  • 15.
    15 What is aday in the life look like for a DBA who has performance issues? Tuning the Database Complex Averages Anscombe's Quartet I II III IV x y x y x y x y Average 9 7.5 9 7.5 9 7.5 9 7.5 Standard Deviation 3.31 2.03 3.31 2.03 3.31 2.03 3.31 2.03 Linear Regression 1.33 1.33 1.33 1.33
  • 16.
    16 LOAD Top Activity SQL Events Sessions GetDetails Click here Max CPU (yard stick) How Can We Open the Black Box?
  • 17.
    17 How do weget our data? Database User 1 User 2 User 3 User 4 Active Sessions = 1 2 3 4 Graph represents # of sessions active, but also represents amount of time active in the database For Every Active Session there is a user (or application) waiting Sample v$session : fast and light weight We collect • session • SQL • State (CPU, IO, LOCK, Other Waits) • and many other columns Multi-dimensional fine grain data. Difficult to use w/o graphics
  • 18.
    18 AWR collects 06/05/08 25 34 36 38 45 F1qcyh20550cf fj6gjgsshtxyx 0cjsxw5ndqdbc 8t8as9usk11qw dr1rkrznhh95b 10dkqv3kr8xa5 38zhkf4jdyff4 298wmz1kxjs1m CPU Enq: TX– row lock contention SQL*Net break/reset to client db file scattered read db file sequential read SQL (v$sqlstats) Sessions (v$sesstat) Waits (v$system_event) Which session executed which SQL? Which SQL was block on a wait event ? AWR ( and Statspack) Can’t answer! Old Method
  • 19.
    19 After sampling Multi-dimensional 06/05/08 2534 36 38 45 63 65 87 F1qcyh20550cf fj6gjgsshtxyx 0cjsxw5ndqdbc 8t8as9usk11qw dr1rkrznhh95b 10dkqv3kr8xa5 38zhkf4jdyff4 298wmz1kxjs1m CPU Enq: TX – row lock contention SQ L*N et break/reset to client db file scattered read db file sequential read IO Application SQL Sessions W aits Statistics Service Scott System User Program Sys Sqlplus SQL*Forms GL OE Package Procedure Plan Child # Module Action File, block, object p1, p2, p3 Sampling V$SESSION
  • 20.
    20 DB Optimizer showsit all Copyright 2006 Kyle Hailey
  • 21.
    21 21 10/30/10 Average Active Sessions(AAS) Use CPU count as yardstick:  AAS < 1 Database is not blocked  AAS ~= 0 Database basically idle Problems are in the APP not DB  AAS >> # of CPUS There is a bottleneck
  • 22.
    22 Interpreting the LoadChart 1. Application – Code inefficient? 1. Database – Configured correctly? 1. Machine – Is the machine undersized? 1. SQL – Inefficient SQL? Chart answers this question and assists with solutions Who’s problem is it ?
  • 23.
    23 1. Application Issues insertinto foo values ('a'); commit; insert into foo values ('a'); commit; insert into foo values ('a'); commit; insert into foo values ('a'); commit; insert into foo values ('a'); commit; insert into foo values ('a'); commit; insert into foo values ('a'); commit; 4 concurrent sessions run :
  • 24.
    24 2. Database issue begin fori in 1..1000 loop insert into foo values ('a'); end loop; end; / Commit; 4 concurrent sessions run :
  • 25.
  • 26.
  • 27.
    27 Tuning SQL Complexity PARSINGIN CURSOR #2 len=53 dep=0 uid=61 oct=3 lid=61 tim=1151519905950403 hv=2296704914 ad='4e50010c' SELECT 'Hello, world; today is ' || SYSDATE FROM dual END OF STMT PARSE #2:c=4000,e=1540,p=0,cr=0,cu=0,mis=1,r=0,dep=0,og=1, tim=1151519905950397 BINDS #2: EXEC #2:c=0,e=58,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=1,tim=1 151519906034782 WAIT #2: nam='SQL*Net message to client' ela= 2 driver id=1650815232 #bytes=1 p3=0 obj#=-1 tim=1151519906034809 FETCH #2:c=0,e=29,p=0,cr=0,cu=0,mis=0,r=1,dep=0,og=1,tim=1 151519906034864 WAIT #2: nam='SQL*Net message from client' ela= 215 driver id=1650815232 #bytes=1 p3=0 obj#=-1 tim=1151519906035133 FETCH #2:c=0,e=1,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=0,tim=11 51519906035165 WAIT #2: nam='SQL*Net message to client' ela= 1 driver id=1650815232 #bytes=1 p3=0 obj#=-1 tim=1151519906035188 WAIT #2: nam='SQL*Net message from client' ela= 192 driver id=1650815232 #bytes=1 p3=0 obj#=-1 tim=1151519906035400 STAT #2 id=1 cnt=1 pid=0 pos=1 obj=0 op='FAST DUAL (cr=0 pr=0 pw=0 time=3 us) -------------------------------------------------------------------------------------------- | Id | Operation | Name | Starts | E-Rows | A-Rows | -------------------------------------------------------------------------------------------- - | 1 | HASH GROUP BY | | 1 | 1 | 1 | |* 2 | FILTER | | 1 | | 1909 | |* 3 | TABLE ACCESS BY INDEX ROWID | PS_RETROPAYPGM_TBL | 1 | 1 | 3413 | | 4 | NESTED LOOPS | | 1 | 165 | 6827 | |* 5 | HASH JOIN | | 1 | 165 | 3413 | |* 6 | HASH JOIN | | 1 | 165 | 3624 | | 7 | TABLE ACCESS BY INDEX ROWID | WB_JOB | 1 | 242 | 2895 | | 8 | NESTED LOOPS | | 1 | 233 | 2897 | | 9 | TABLE ACCESS BY INDEX ROWID| PS_PAY_CALENDAR | 1 | 1 | 1 | |* 10 | INDEX RANGE SCAN | PS0PAY_CALENDAR | 1 | 1 | 1 | |* 11 | INDEX RANGE SCAN | WBBJOB_B | 1 | 286 | 2895 | |* 12 | TABLE ACCESS FULL | WB_RETROPAY_EARNS | 1 | 27456 | 122K| | 13 | TABLE ACCESS FULL | PS_RETROPAY_RQST | 1 | 13679 | 13679 | |* 14 | INDEX RANGE SCAN | PS#RETROPAYPGM_TBL | 3413 | 1 | 3413 | | 15 | SORT AGGREGATE | | 1791 | 1 | 1791 | | 16 | FIRST ROW | | 1791 | 1 | 1579 | |* 17 | INDEX RANGE SCAN (MIN/MAX) | WB_JOB_F | 1791 | 1 | 1579 | | 18 | SORT AGGREGATE | | 1539 | 1 | 1539 | | 19 | FIRST ROW | | 1539 | 1 | 1539 | |* 20 | INDEX RANGE SCAN (MIN/MAX) | WB_JOB_G | 1539 | 1 | 1539 Trace file Execution Plan Complexity !
  • 28.
  • 29.
    29 Tuning SQL • Indexes –Missing indexes • Table Stats – Stale statistics • Histograms – Recommendations • Execution Plan – DB Optimizer automatically finds better plans • Visual SQL Tuning diagrams
  • 30.
    30 VST Steps 1. Tables –drawn as nodes 1. Joins – drawn as connector lines 1. Filters – mark on each table with filter in where clause 30 10/30/10
  • 31.
    31 How to VST:Tables and Joins SELECT C.Phone_Number, C.Honorific, C.First_Name, C.Last_Name, C.Suffix, C.Address_ID, A.Address_ID, A.Street_Address_Line1, A.Street_Address_Line2, A.City_Name, A.State_Abbreviation, A.ZIP_Code, OD.Deferred_Shipment_Date, OD.Item_Count, ODT.Text, OT.Text, P.Product_Description, S.Shipment_Date FROM Orders O, Order_Details OD, Products P, Customers C, Shipments S, Addresses A, Code_Translations ODT, Code_Translations OT WHERE UPPER(C.Last_Name) LIKE :Last_Name||'%' AND UPPER(C.First_Name) LIKE :First_Name||'%' AND OD.Order_ID = O.Order_ID AND O.Customer_ID = C.Customer_ID AND OD.Product_ID = P.Product_ID(+) AND OD.Shipment_ID = S.Shipment_ID(+) AND S.Address_ID = A.Address_ID(+) AND O.Status_Code = OT.Code AND OT.Code_Type = 'ORDER_STATUS' AND OD.Status_Code = ODT.Code AND ODT.Code_Type = 'ORDER_DETAIL_STATUS' AND O.Order_Date > :Now - 366 ORDER BY C.Customer_ID, O.Order_ID DESC, S.Shipment_ID, OD.Order_Detail_ID; Copyright 2006 Kyle Hailey Tables Orders O, Order_Details OD, Products P, Customers C, Shipments S, Addresses A, Code_Translations ODT, Code_Translations OT Joins OD.Order_ID = O.Order_ID O.Customer_ID = C.Customer_ID OD.Product_ID = P.Product_ID(+) OD.Shipment_ID = S.Shipment_ID(+) S.Address_ID = A.Address_ID(+) O.Status_Code = OT.Code OD.Status_Code = ODT.Code Dan Tow – SQL TUNING Filters WHERE UPPER(C.Last_Name) LIKE :Last_Name||'%' AND UPPER(C.First_Name) LIKE :First_Name||'%' AND OT.Code_Type = 'ORDER_STATUS' AND O.Order_Date > :Now – 366 AND ODT.Code_Type = 'ORDER_DETAIL_STATUS'
  • 32.
    32 Layout tables andconnections Tables Orders O, Order_Details OD, Products P, Customers C, Shipments S, Addresses A, Code_Translations ODT, Code_Translations OT Joins OD.Order_ID = O.Order_ID O.Customer_ID = C.Customer_ID OD.Product_ID = P.Product_ID(+) OD.Shipment_ID = S.Shipment_ID(+) S.Address_ID = A.Address_ID(+) O.Status_Code = OT.Code OD.Status_Code = ODT.Code F S A OD P ODT C OT O
  • 33.
    33 Unstructured Copyright 2006 Kyle Hailey Joins OD.Order_ID= O.Order_ID O.Customer_ID = C.Customer_ID OD.Product_ID = P.Product_ID(+) OD.Shipment_ID = S.Shipment_ID(+) S.Address_ID = A.Address_ID(+) O.Status_Code = OT.Code OD.Status_Code = ODT.Code Neater, but can you do anything with it? What’s the optimal execution path? S A OD P ODT C OT O
  • 34.
    34 Parents and Children Copyright2006 Kyle Hailey Joins OD.Order_ID = O.Order_ID O.Customer_ID = C.Customer_ID OD.Product_ID = P.Product_ID(+) OD.Shipment_ID = S.Shipment_ID(+) S.Address_ID = A.Address_ID(+) O.Status_Code = OT.Code OD.Status_Code = ODT.Code ODT OD P C A O S OT Primary Key (unique index) No index or non-unique Master Detail Structure the tree
  • 35.
    35 VST – filtersand best path Parent Child Parent Child Concept: 1. Start at most selective filter 2. Join down first, before joining upwards Filters WHERE UPPER(C.Last_Name) LIKE :Last_Name||'%' AND UPPER(C.First_Name) LIKE :First_Name||'%' AND OT.Code_Type = 'ORDER_STATUS‘ AND ODT.Code_Type = 'ORDER_DETAIL_STATUS' AND O.Order_Date > :Now – 366  Filters help determine best path F F 30% 0.02% 100% * (select count(*) from TAB where condition) --------------------------- (select count(*) from Tab) S A OD P ODT C OT O
  • 36.
    36 F F 30% 0.02% S A OD P ODT C OT O VST– best path Note: Oracle only joins in one table to the previous result set F F
  • 37.
    37 Cartesian SELECT A.BROKER_ID BROKER_ID, A.BROKER_LAST_NAME BROKER_LAST_NAME, A.BROKER_FIRST_NAMEBROKER_FIRST_NAME, A.YEARS_WITH_FIRM YEARS_WITH_FIRM, C.OFFICE_NAME OFFICE_NAME, SUM (B.BROKER_COMMISSION) TOTAL_COMMISSIONS FROM BROKER A, CLIENT_TRANSACTION B, OFFICE_LOCATION C, INVESTMENT I WHERE A.BROKER_ID = B.BROKER_ID AND A.OFFICE_LOCATION_ID = C.OFFICE_LOCATION_ID GROUP BY A.BROKER_ID, A.BROKER_LAST_NAME, A.BROKER_FIRST_NAME, A.YEARS_WITH_FIRM, C.OFFICE_NAME;
  • 38.
    38 Implied Cartesian select c.client_first_name, c.client_last_name, ct.action,ct.price, b.broker_last_name, b.broker_first_name, o.office_name from client_transaction ct, client c, broker b, office_location o where ct.price > 100 and b.broker_id=ct.broker_id and c.broker_id = b.broker_id and o.office_location_id = b.office_location_id
  • 39.
    39 Diagram work forMany to One What about many to many?
  • 40.
    40 Unstructured Copyright 2006 Kyle Hailey Joins OD.Order_ID= O.Order_ID O.Customer_ID = C.Customer_ID OD.Product_ID = P.Product_ID(+) OD.Shipment_ID = S.Shipment_ID(+) S.Address_ID = A.Address_ID(+) O.Status_Code = OT.Code OD.Status_Code = ODT.Code S A OD P ODT C OT O S A OD P ODT C OT O v v v v v v v v v v v v v v F F 30% 0.02%
  • 41.
    41 Many-to-One vs Many-to-Many PredicateFilter 1 2 3 B -> C -> A go to A or C? Now what?
  • 42.
    42 Adding Constraints SELECT COUNT(*) FROM b, c, a WHERE b.val2 = 100 AND a.val1 = b.id AND b.val1 = c.id; 58 logical reads alter table c add constraint c_pk unique (id); alter table b add constraint b_pk unique (id); 7 logical reads
  • 43.
    43 Join sizes Join Sizes Usethe Join sizes to determine path of least resistance
  • 44.
    44 Look at 3queries • Query 1 runs more than 24 hours • Query 2 outer joins and scalar subqueries • Query 3 create path not available to Oracle
  • 45.
    45 Query 1 :Over 24 hours to run SELECT A0.zuchinis, A0.brocoli, C0.Oranges FROM ( SELECT A1.planted_date, A1.pears, A1.zuchinis, A1.brocoli FROM FOO.A A1, ( SELECT zuchinis, brocoli FROM FOO.A A2 WHERE pears = 'M' AND planted_date + 0 >= ADD_MONTHS ((SELECT MAX (planted_date) FROM FOO.B B1 WHERE pears = 'M' ), - 11) GROUP BY zuchinis, brocoli HAVING COUNT (*) = 12 ) i2 WHERE A1.planted_date = (SELECT MAX (planted_date) FROM FOO.B B2 WHERE pears = 'M' ) AND A1.pears = 'M' AND A1.zuchinis = i2.zuchinis (+) AND A1.brocoli = i2.brocoli (+) UNION SELECT A4.planted_date, A4.pears,A4.zuchinis,A4.brocoli FROM FOO.A A4 WHERE A4.planted_date >'01-OCT-08' and A4.planted_date <'03-OCT-08' AND A4.pears = 'D' AND A4.green_beans = '1' AND NOT EXISTS (SELECT * FROM FOO.A A5 WHERE pears = 'M' AND planted_date = (SELECT MAX (planted_date) FROM FOO.B B3 WHERE pears = 'M' ) AND A4.zuchinis = A5.zuchinis AND A4.brocoli = A5.brocoli) ) b, FOO.A A0, FOO.C C0, FOO.D D0, FOO.E E0 WHERE A0.planted_date >'01-OCT-08' and A0.planted_date <'03-OCT-08' AND A0.pears = 'D' AND A0.green_beans = '1' AND A0.zuchinis = b.zuchinis AND A0.brocoli = b.brocoli AND A0.planted_date = C0.planted_date AND A0.pears = C0.pears AND A0.zuchinis = C0.zuchinis AND A0.brocoli = C0.brocoli AND A0.planted_date = D0.planted_date AND A0.pears = D0.pears AND A0.harvest_size = D0.harvest_size AND C0.Oranges = D0.Oranges AND C0.apples = D0.apples AND (D0.lemons = 0 OR D0.lemons IS NULL) AND A0.planted_date = E0.planted_date AND A0.pears = E0.pears AND A0.harvest_size = E0.harvest_size AND C0.Oranges = E0.Oranges AND C0.apples = E0.apples AND (E0.lemons = 0 OR E0.lemons IS NULL) ORDER BY A0.zuchinis, A0.brocoli;
  • 46.
  • 47.
  • 48.
    48 Comparing Plans :24 hours to 5 mins NL NL NL NL HJ HJ NL NL
  • 49.
    49 Q2 49 10/30/10 SELECT CASEWHEN M.NYC IS NULL THEN (SELECT /*+ qb_name(qb1) */ MAX (Kona) FROM foo.F WHERE harvest_date = to_date('08/10/2008','dd/mm/yyyy') AND Argentina = TRIM ('D') AND Norway = F_OUTER.Norway ELSE M.NYC END AS NYC, CASE WHEN F_OUTER.Perth IS NULL THEN NULL ELSE (SELECT /*+ qb_name(qb2) */ Georgia FROM foo.P WHERE harvest_date = to_date('08/10/2008','dd/mm/yyyy') AND Argentina = TRIM ('D') AND Paris = F_OUTER.Perth) END AS richard, CASE WHEN F_OUTER.Aruba IS NULL THEN NULL ELSE (SELECT /*+ qb_name(qb3) */ Georgia FROM foo.P WHERE harvest_date = to_date('08/10/2008','dd/mm/yyyy') AND Argentina = TRIM ('D') AND Paris = F_OUTER.Aruba) END AS Jody, CASE WHEN F_OUTER.Portland IS NULL THEN NULL ELSE (SELECT /*+ qb_name(qb4) */ Georgia FROM foo.P WHERE harvest_date = to_date('08/10/2008','dd/mm/yyyy') AND Argentina = TRIM ('D') AND Paris = F_OUTER.Portland) END AS Tom FROM foo.F F_OUTER, foo.M , foo.J , foo.N , (SELECT /*+ qb_name(qb5) */ H.SF, Oregon, H.Haiti, K.Bermuda, L.Denmark FROM (foo.H LEFT OUTER JOIN foo.K ON H.harvest_date = K.harvest_date AND H.Argentina = K.Argentina AND H.SF = K.SF AND K.Dallas = '001') LEFT OUTER JOIN FOo.L ON H.harvest_date = L.harvest_date AND H.Argentina = L.Argentina AND H.SF = L.SF WHERE H.harvest_date = to_date('08/10/2008','dd/mm/yyyy') AND H.Argentina = TRIM ('D')) extra WHERE F_OUTER.harvest_date = M.harvest_date(+) AND F_OUTER.Argentina = M.Argentina(+) AND F_OUTER.Norway = M.Norway(+) AND M.Norway(+) = M.Texas(+) AND F_OUTER.harvest_date = to_date('08/10/2008','dd/mm/yyyy') AND F_OUTER.Argentina = TRIM ('D') AND M.harvest_date(+) = to_date('08/10/2008','dd/mm/yyyy') AND M.Argentina(+) = TRIM ('D') AND F_OUTER.Norway = F_OUTER.Hawaii AND F_OUTER.harvest_date = J.harvest_date(+) AND F_OUTER.Argentina = J.Argentina(+) AND F_OUTER.Norway = J.Texas(+) AND J.harvest_date(+) = to_date('08/10/2008','dd/mm/yyyy') AND J.Argentina(+) = TRIM ('D') AND F_OUTER.Iraq = extra.SF(+) AND F_OUTER.harvest_date = N.harvest_date(+) Two important qualities: • All outer joins to F_OUTER • 4 subqueries in select select
  • 50.
  • 51.
    51 Q2 51 10/30/10 The subqueriesin the select clause look like select CASE WHEN F.f1 IS NULL THEN NULL ELSE (SELECT X.f2 FROM X WHERE code_vl = F.f1) END AS f0 from F; and should be merged into the query like: select CASE WHEN F.f1 IS NULL THEN NULL ELSE ( X.f2) END AS f0 from F , X where code_vl(+) = F.f1; select CASE WHEN F.f1 IS NULL THEN NULL ELSE (SELECT X.f2 FROM X WHERE code_vl = F.f2) END AS f0 from F; select CASE WHEN F.f1 IS NULL THEN NULL ELSE ( X.f2) END AS f0 from F , X where code_vl(+) = F.f1;
  • 52.
    52 Q3 52 10/30/10 SELECT DISTINCT* FROM FOO.a a, FOO.c c, FOO.d d, FOO.g g WHERE a.planted_date > '01-OCT-08' AND a.planted_date < '03-OCT-08' AND a.pears = 'D' AND a.green_beans = '1' AND a.planted_date = c.planted_date AND a.pears = c.pears AND a.zuchinis = c.zuchinis AND a.brocoli = c.brocoli AND a.planted_date = d.planted_date AND a.pears = d.pears AND a.harvest_size = d.harvest_size AND c.oranges = d.oranges AND c.apples = d.apples AND (d.lemons = 0 OR d.lemons IS NULL) AND a.planted_date = g.planted_date AND a.pears = g.pears AND a.harvest_size = g.harvest_size AND c.oranges = g.oranges AND c.apples = g.apples AND (g.lemons = 0 OR g.lemons IS NULL) AND a.zuchinis = '0236' AND d.apples = g.apples AND d.oranges = g.oranges ORDER BY a.zuchinis, a.brocoli;
  • 53.
  • 54.
    54 Q3 54 10/30/10 SELECT *FROM ( SELECT /*+ NO_MERGE */ c.apples, c.oranges, a.harvest_size FROM a, c WHERE a.planted_date = TO_DATE ('02/10/2008', 'dd/mm/yyyy') AND a.pears = 'D' AND a.green_beans = '1' AND a.planted_date = c.planted_date AND a.pears = c.pears AND a.zuchinis = c.zuchinis AND a.brocoli = c.brocoli AND a.zuchinis = '0236' ) X, ( SELECT /*+ NO_MERGE */ d.apples, d.oranges, d.harvest_size FROM d, g WHERE d.planted_date = TO_DATE ('02/10/2008', 'dd/mm/yyyy') AND g.planted_date = TO_DATE ('02/10/2008', 'dd/mm/yyyy') AND g.apples = d.apples AND d.oranges = g.oranges AND d.pears = 'D' AND g.pears = 'D' AND g.pears = d.pears AND g.harvest_size = d.harvest_size AND (d.lemons = 0 OR d.lemons IS NULL) AND (g.lemons = 0 OR g.lemons IS NULL) ) Y WHERE X.oranges = Y.oranges AND X.apples = Y.apples AND X.harvest_size = Y.harvest_size; This final version runs in elapsed 0.33 secs and 12K logical reads down from an original elapsed 4.5 secs and 1M logical reads
  • 55.
    55 What tools touse to create VST diagrams? • Paper – Modifications messy and difficult • PowerPoint – Can move things around – Sticky connectors – Easy to modify • DB Optimizer – Automatic and still modifiable
  • 56.
    56 Visual SQL Tuning(VST) Table Sizes Join Sizes Filter Ratios Exists
  • 57.
    57 VST Steps Summary 1.Diagramtables 2.Draw connectors for each join 3.Calculate filter ratios 4.Find the table sizes 5.Calculate two table join sizes Execution Path • Start at the most selective join filter • Join to keep the running result set size small Many to Many relationships = problems 57 10/30/10 Parent Child Parent Child F F 30% 0.02% S A OD P ODT C OT O
  • 58.
    58 Copyright 2006 Kyle Hailey DoYou Want? Engineering Data? Engineering Data?
  • 59.
    59 Copyright 2006 Kyle Hailey DoYou Want? Pretty Pictures Pretty Pictures
  • 60.
    60 Copyright 2006 Kyle Hailey DoYou Want? Clean and Clear Clean and Clear ? ? ? ? ? ?
  • 61.
    61 Copyright 2006 Kyle Hailey ImagineTrying to Drive your Car And is updated once and hour And is updated once and hour Or would you like it to look … Would you want your dashboard to look like :
  • 62.
  • 63.
    63 Summary 1.Database - AAS –Profile database – Use wait interface and graphics – Identify machine, application, database or SQL 1.SQL - VST – Indexes, stats, execution path – Visual SQL Tuning
  • 64.
    64 Bibliography http://www.simple-talk.com/sql/performance/designing-efficient-sql-a-visual-approach/ Refactoring SQL Applications– Stephane Faroult Troubleshooting Oracle Performance – Christian Antognini SQL Tuning – Dan Tow Cost-Based Oracle Fundamentals – Jonathan Lewis
  • 65.
    65 Visual SQL Tuning(VST) Industry Exclusive
  • 66.
    66 DB Optimizer XEKey Features Profile Tune Load Test SQL IDE
  • 67.
    67 DB Optimizer XEKey Features Profile Tune Load Test SQL IDE
  • 68.
    68 DB Optimizer XEKey Features Profile Tune Load Test SQL IDE
  • 69.
    69 DB Optimizer XEKey Features Profile Tune Load Test SQL IDE
  • 70.
    70 Review of Functionality Profile Tune LoadTest SQL IDE Only tool on the market with these features integrated
  • 71.
  • 72.
    72 Appendix • Diagramming – Simplejoins and inline views – Outer joins – Exists/not exists (in/not in) – Correlated aggregate sub-queries • Hints – LEADING – USE_NL – USE_HASH – INDEX – NO_MERGE • Execution Order • Machine Health • VST method with Statistics
  • 73.
    73 Basic Joins Many to singlevalue Cartesian Cartesian Safe Dangerous M
  • 74.
    74 Simple queries andsub-queries 74 A B select * from A, B where b.f1=a.f1 Non-correlated sub-query select * from A,(select b.f1 from B) c where c.f1=a.f1 Simple join select * from A where a.f1 = (select max(b.f1) from B ) Special case: non correlated sub-query returns one row A B A B A B Unique index on b.f1 No unique indexes Unique index on Both b.f1 and a.f1 M
  • 75.
  • 76.
    76 Correlated scalar sub-queries 7610/30/10 Correlated aggregate subquery: select ename from emp a where a.sal > (select avg(sal) from emp b where a.deptno=b.deptno) A B Scalar Subqueries select ename, (select avg(sal) from emp b where a.deptno=b.deptno) from emp a;
  • 77.
    77 Exists and NotIn 77 10/30/10 SELECT d.* FROM dept d WHERE exists ( SELECT null FROM emp e WHERE e.deptno=d.deptno); SELECT d.* FROM dept d WHERE d.deptno in ( SELECT deptno FROM emp e ); select distinct d.* from dept1 d ,emp e where e.deptno = d.deptno; SELECT d.* FROM dept d WHERE not exists ( SELECT null FROM emp e WHERE e.deptno=d.deptno); SELECT d.* FROM dept d WHERE d.deptno not in ( SELECT deptno FROM emp e where e.deptno is not null ) or d.deptno is null; select d.* from dept1 d left outer join emp e on e.deptno = d.deptno where e.deptno is null; E N
  • 78.
    78 HINTS • ORDERED -good on 9i • Leading(tab_alias , table_alias … ) – 10g format • USE_NL(table_alias) – Inner Table (not driving) • USE_HASH(table_alias) – 2cd table, probe into • INDEX(tab_alias index_name) • NO_MERGE Oracle first decides join order then join type (example http://www.adp-gmbh.ch/blog/2008/01/17.php) 78 10/30/10
  • 79.
    79 Visual SQL Tuning(VST) diagrams 79 10/30/10 HJ NL i io HJ NL Makes comparison of execution plans easy Better yet, it will help us find the best execution path
  • 80.
  • 81.
    81 i MJ HJ Bad Plans vsGood Plan 81 10/30/10 HJ NL
  • 82.
    82 Comparing Plans :24 hours to 5 mins NL NL NL NL HJ HJ NL NL
  • 83.
    83 When to Tune 1.Machine a) CPU • Response times skewed • 100% CPU might be fine • Users wait in queue (run queue) => machine underpowered a) Memory • Paging • Wait times skewed (ex : latch free) • Erratic response times ( ex : ls ) 1. Oracle 1) Waits > CPU ? • tune waits 1) CPU > 100% ? • tune top CPU SQL 1) Else • It’s the application Oracle Load (AAS) Top Session Top Wait Top SQL SQL Detail Session Detail File Detail Object Detail Wait Detail Host AAS > #CPU AAS > 1 Waits > CPU CPU > Waits CPU Memory
  • 84.
    84 Machine Make sure themachine is healthy before tuning Oracle • CPU => use run queue, < 2 * #CPU • Memory => page out VMSTAT
  • 85.
    85 Summary 1.Machine - vmstat –Memory, CPU (we can see IO response in Oracle) 1.Database - AAS – Use wait interface and graphics – Identify machine, application, database or SQL 1.SQL - VST – Indexes, stats, execution path – Visual SQL Tuning
  • 86.
    86 Solution to Manyto Many Table Sizes Filter Ratios Join Sizes
  • 87.
    87 VST Steps Objects: 1. Tables: drawn as nodes 2. Joins : drawn as connector lines 3. Filters : mark on each table with filter in where clause Statistics: 1.Table sizes 2. Join sizes 3. Calculate filter percentages – filter ratio = number of rows returned with filter / number of rows 87 10/30/10
  • 88.
    88 Google Directions • Comparetwo sets of directions • (optionally show graphical set of directions) • Show map • Show map with traffic flow, read, yellow, green
  • 89.
    89 How Can WeOpen the Black Box? OEM ASHMON/SASH DB Optimizer •Powerful - Identifies issues quickly and powerfully •Interactive - Allows exploring the data •Easy - Understandable by everyone, DBA, Dev and Managers !
  • 90.
    90 Copyright 2006 Kyle Hailey Clearer 1.Include successes 2. Remove Irrelevant 3. Normalize same temp 4. Scale known vs unknown 55 65 75 60 70 80 4 8 12 4 8 12 30 40 50 35 45 X