Oracle 23ai New Feature: Automatic SQL Transpiler
We know that both SQL and PL/SQL have their own execution engines. Every time we call a user defined PL/SQL function from a SQL query, or execute an SQL query in within a PL/SQL code, there is a context switch between the SQL and PL/SQL engines. The cost based optimizer can spend some time to do these context switches. Usually this time is not that significant, but if the number of context switches takes place repeatedly, then these calls might incur significant overhead because the PL/SQL runtime must be invoked. Moreover, since the PL/SQL code is compiled code, the cost based optimizer does not know what is going on inside the PL/SQL code. This messes up the execution plan of the query.
The Automatic SQL Transpiler addresses this problem. If Automatic Transpiler is switched on, then the transpiler will attempt to transform the PL/SQL functions, wherever possible, within SQL into SQL expressions, without user intervention. If successful, this will avoid the need for context switches and allows the optimizer to come up with a better execution plan.
Automatic SQL Transpiler can be switched ON/OFF using the sql_transpiler parameter. By default it is switched OFF.
SQL> show parameter sql_transpiler
NAME TYPE VALUE
-------------- ------ -----
sql_transpiler string OFF
Let’s see with an example. We will work with the employees table.
SQL> Select * from hr.employees;
EMPLOYEE_ID FIRST_NAME LAST_NAME EMAIL PHONE_NUMBER HIRE_DATE JOB_ID SALARY COMMISSION_PCT MANAGER_ID DEPARTMENT_ID
______________ ______________ ____________ ___________ _______________ ____________ _____________ _________ _________________ _____________ ________________
100 Steven King SKING 515.123.4567 17-JUN-03 AD_PRES 24000 90
101 Neena Kochhar NKOCHHAR 515.123.4568 21-SEP-05 AD_VP 17000 100 90
102 Lex De Haan LDEHAAN 515.123.4569 13-JAN-01 AD_VP 17000 100 90
103 Alexander Hunold AHUNOLD 590.423.4567 03-JAN-06 IT_PROG 9000 102 60
104 Bruce Ernst BERNST 590.423.4568 21-MAY-07 IT_PROG 6000 103 60
105 David Austin DAUSTIN 590.423.4569 25-JUN-05 IT_PROG 4800 103 60
106 Valli Pataballa VPATABAL 590.423.4560 05-FEB-06 IT_PROG 4800 103 60
107 Diana Lorentz DLORENTZ 590.423.5567 07-FEB-07 IT_PROG 4200 103 60
108 Nancy Greenberg NGREENBE 515.124.4569 17-AUG-02 FI_MGR 12008 101 100
109 Daniel Faviet DFAVIET 515.124.4169 16-AUG-02 FI_ACCOUNT 9000 108 100
110 John Chen JCHEN 515.124.4269 28-SEP-05 FI_ACCOUNT 8200 108 100
111 Ismael Sciarra ISCIARRA 515.124.4369 30-SEP-05 FI_ACCOUNT 7700 108 100
112 Jose Manuel Urman JMURMAN 515.124.4469 07-MAR-06 FI_ACCOUNT 7800 108 100
113 Luis Popp LPOPP 515.124.4567 07-DEC-07 FI_ACCOUNT 6900 108 100
114 Den Raphaely DRAPHEAL 515.127.4561 07-DEC-02 PU_MAN 11000 100 30
115 Alexander Khoo AKHOO 515.127.4562 18-MAY-03 PU_CLERK 3100 114 30
116 Shelli Baida SBAIDA 515.127.4563 24-DEC-05 PU_CLERK 2900 114 30
117 Sigal Tobias STOBIAS 515.127.4564 24-JUL-05 PU_CLERK 2800 114 30
118 Guy Himuro GHIMURO 515.127.4565 15-NOV-06 PU_CLERK 2600 114 30
119 Karen Colmenares KCOLMENA 515.127.4566 10-AUG-07 PU_CLERK 2500 114 30
120 Matthew Weiss MWEISS 650.123.1234 18-JUL-04 ST_MAN 8000 100 50
121 Adam Fripp AFRIPP 650.123.2234 10-APR-05 ST_MAN 8200 100 50
122 Payam Kaufling PKAUFLIN 650.123.3234 01-MAY-03 ST_MAN 7900 100 50
123 Shanta Vollman SVOLLMAN 650.123.4234 10-OCT-05 ST_MAN 6500 100 50
124 Kevin Mourgos KMOURGOS 650.123.5234 16-NOV-07 ST_MAN 5800 100 50
125 Julia Nayer JNAYER 650.124.1214 16-JUL-05 ST_CLERK 3200 120 50
126 Irene Mikkilineni IMIKKILI 650.124.1224 28-SEP-06 ST_CLERK 2700 120 50
127 James Landry JLANDRY 650.124.1334 14-JAN-07 ST_CLERK 2400 120 50
128 Steven Markle SMARKLE 650.124.1434 08-MAR-08 ST_CLERK 2200 120 50
129 Laura Bissot LBISSOT 650.124.5234 20-AUG-05 ST_CLERK 3300 121 50
130 Mozhe Atkinson MATKINSO 650.124.6234 30-OCT-05 ST_CLERK 2800 121 50
131 James Marlow JAMRLOW 650.124.7234 16-FEB-05 ST_CLERK 2500 121 50
132 TJ Olson TJOLSON 650.124.8234 10-APR-07 ST_CLERK 2100 121 50
133 Jason Mallin JMALLIN 650.127.1934 14-JUN-04 ST_CLERK 3300 122 50
134 Michael Rogers MROGERS 650.127.1834 26-AUG-06 ST_CLERK 2900 122 50
135 Ki Gee KGEE 650.127.1734 12-DEC-07 ST_CLERK 2400 122 50
136 Hazel Philtanker HPHILTAN 650.127.1634 06-FEB-08 ST_CLERK 2200 122 50
137 Renske Ladwig RLADWIG 650.121.1234 14-JUL-03 ST_CLERK 3600 123 50
138 Stephen Stiles SSTILES 650.121.2034 26-OCT-05 ST_CLERK 3200 123 50
139 John Seo JSEO 650.121.2019 12-FEB-06 ST_CLERK 2700 123 50
140 Joshua Patel JPATEL 650.121.1834 06-APR-06 ST_CLERK 2500 123 50
141 Trenna Rajs TRAJS 650.121.8009 17-OCT-03 ST_CLERK 3500 124 50
142 Curtis Davies CDAVIES 650.121.2994 29-JAN-05 ST_CLERK 3100 124 50
143 Randall Matos RMATOS 650.121.2874 15-MAR-06 ST_CLERK 2600 124 50
144 Peter Vargas PVARGAS 650.121.2004 09-JUL-06 ST_CLERK 2500 124 50
145 John Russell JRUSSEL 011.44.1344.429268 01-OCT-04 SA_MAN 14000 0.4 100 80
146 Karen Partners KPARTNER 011.44.1344.467268 05-JAN-05 SA_MAN 13500 0.3 100 80
147 Alberto Errazuriz AERRAZUR 011.44.1344.429278 10-MAR-05 SA_MAN 12000 0.3 100 80
148 Gerald Cambrault GCAMBRAU 011.44.1344.619268 15-OCT-07 SA_MAN 11000 0.3 100 80
149 Eleni Zlotkey EZLOTKEY 011.44.1344.429018 29-JAN-08 SA_MAN 10500 0.2 100 80
150 Peter Tucker PTUCKER 011.44.1344.129268 30-JAN-05 SA_REP 10000 0.3 145 80
151 David Bernstein DBERNSTE 011.44.1344.345268 24-MAR-05 SA_REP 9500 0.25 145 80
152 Peter Hall PHALL 011.44.1344.478968 20-AUG-05 SA_REP 9000 0.25 145 80
153 Christopher Olsen COLSEN 011.44.1344.498718 30-MAR-06 SA_REP 8000 0.2 145 80
154 Nanette Cambrault NCAMBRAU 011.44.1344.987668 09-DEC-06 SA_REP 7500 0.2 145 80
155 Oliver Tuvault OTUVAULT 011.44.1344.486508 23-NOV-07 SA_REP 7000 0.15 145 80
156 Janette King JKING 011.44.1345.429268 30-JAN-04 SA_REP 10000 0.35 146 80
157 Patrick Sully PSULLY 011.44.1345.929268 04-MAR-04 SA_REP 9500 0.35 146 80
158 Allan McEwen AMCEWEN 011.44.1345.829268 01-AUG-04 SA_REP 9000 0.35 146 80
159 Lindsey Smith LSMITH 011.44.1345.729268 10-MAR-05 SA_REP 8000 0.3 146 80
160 Louise Doran LDORAN 011.44.1345.629268 15-DEC-05 SA_REP 7500 0.3 146 80
161 Sarath Sewall SSEWALL 011.44.1345.529268 03-NOV-06 SA_REP 7000 0.25 146 80
162 Clara Vishney CVISHNEY 011.44.1346.129268 11-NOV-05 SA_REP 10500 0.25 147 80
163 Danielle Greene DGREENE 011.44.1346.229268 19-MAR-07 SA_REP 9500 0.15 147 80
164 Mattea Marvins MMARVINS 011.44.1346.329268 24-JAN-08 SA_REP 7200 0.1 147 80
165 David Lee DLEE 011.44.1346.529268 23-FEB-08 SA_REP 6800 0.1 147 80
166 Sundar Ande SANDE 011.44.1346.629268 24-MAR-08 SA_REP 6400 0.1 147 80
167 Amit Banda ABANDA 011.44.1346.729268 21-APR-08 SA_REP 6200 0.1 147 80
168 Lisa Ozer LOZER 011.44.1343.929268 11-MAR-05 SA_REP 11500 0.25 148 80
169 Harrison Bloom HBLOOM 011.44.1343.829268 23-MAR-06 SA_REP 10000 0.2 148 80
170 Tayler Fox TFOX 011.44.1343.729268 24-JAN-06 SA_REP 9600 0.2 148 80
171 William Smith WSMITH 011.44.1343.629268 23-FEB-07 SA_REP 7400 0.15 148 80
172 Elizabeth Bates EBATES 011.44.1343.529268 24-MAR-07 SA_REP 7300 0.15 148 80
173 Sundita Kumar SKUMAR 011.44.1343.329268 21-APR-08 SA_REP 6100 0.1 148 80
174 Ellen Abel EABEL 011.44.1644.429267 11-MAY-04 SA_REP 11000 0.3 149 80
175 Alyssa Hutton AHUTTON 011.44.1644.429266 19-MAR-05 SA_REP 8800 0.25 149 80
176 Jonathon Taylor JTAYLOR 011.44.1644.429265 24-MAR-06 SA_REP 8600 0.2 149 80
177 Jack Livingston JLIVINGS 011.44.1644.429264 23-APR-06 SA_REP 8400 0.2 149 80
178 Kimberely Grant KGRANT 011.44.1644.429263 24-MAY-07 SA_REP 7000 0.15 149
179 Charles Johnson CJOHNSON 011.44.1644.429262 04-JAN-08 SA_REP 6200 0.1 149 80
180 Winston Taylor WTAYLOR 650.507.9876 24-JAN-06 SH_CLERK 3200 120 50
181 Jean Fleaur JFLEAUR 650.507.9877 23-FEB-06 SH_CLERK 3100 120 50
182 Martha Sullivan MSULLIVA 650.507.9878 21-JUN-07 SH_CLERK 2500 120 50
183 Girard Geoni GGEONI 650.507.9879 03-FEB-08 SH_CLERK 2800 120 50
184 Nandita Sarchand NSARCHAN 650.509.1876 27-JAN-04 SH_CLERK 4200 121 50
185 Alexis Bull ABULL 650.509.2876 20-FEB-05 SH_CLERK 4100 121 50
186 Julia Dellinger JDELLING 650.509.3876 24-JUN-06 SH_CLERK 3400 121 50
187 Anthony Cabrio ACABRIO 650.509.4876 07-FEB-07 SH_CLERK 3000 121 50
188 Kelly Chung KCHUNG 650.505.1876 14-JUN-05 SH_CLERK 3800 122 50
189 Jennifer Dilly JDILLY 650.505.2876 13-AUG-05 SH_CLERK 3600 122 50
190 Timothy Gates TGATES 650.505.3876 11-JUL-06 SH_CLERK 2900 122 50
191 Randall Perkins RPERKINS 650.505.4876 19-DEC-07 SH_CLERK 2500 122 50
192 Sarah Bell SBELL 650.501.1876 04-FEB-04 SH_CLERK 4000 123 50
193 Britney Everett BEVERETT 650.501.2876 03-MAR-05 SH_CLERK 3900 123 50
194 Samuel McCain SMCCAIN 650.501.3876 01-JUL-06 SH_CLERK 3200 123 50
195 Vance Jones VJONES 650.501.4876 17-MAR-07 SH_CLERK 2800 123 50
196 Alana Walsh AWALSH 650.507.9811 24-APR-06 SH_CLERK 3100 124 50
197 Kevin Feeney KFEENEY 650.507.9822 23-MAY-06 SH_CLERK 3000 124 50
198 Donald OConnell DOCONNEL 650.507.9833 21-JUN-07 SH_CLERK 2600 124 50
199 Douglas Grant DGRANT 650.507.9844 13-JAN-08 SH_CLERK 2600 124 50
200 Jennifer Whalen JWHALEN 515.123.4444 17-SEP-03 AD_ASST 4400 101 10
201 Michael Hartstein MHARTSTE 515.123.5555 17-FEB-04 MK_MAN 13000 100 20
202 Pat Fay PFAY 603.123.6666 17-AUG-05 MK_REP 6000 201 20
203 Susan Mavris SMAVRIS 515.123.7777 07-JUN-02 HR_REP 6500 101 40
204 Hermann Baer HBAER 515.123.8888 07-JUN-02 PR_REP 10000 101 70
205 Shelley Higgins SHIGGINS 515.123.8080 07-JUN-02 AC_MGR 12008 101 110
206 William Gietz WGIETZ 515.123.8181 07-JUN-02 AC_ACCOUNT 8300 205 110
90000 Rishin Mitra RMitra 12345 07-JUN-02 IT_PROG 12008 210
99999 Darryl G DG 67890 07-JUN-02 IT_PROG 12008 300
109 rows selected.
We create a function to calculate the total salary[ salary + (salary * commission_pct) ].
SQL> CREATE OR REPLACE FUNCTION total_sal(salary NUMBER, commission_pct NUMBER)
RETURN NUMBER
AS
BEGIN
RETURN salary + ((NVL(commission_pct, 0) * salary)/100);
END;
/
Function TOTAL_SAL compiled
Elapsed: 00:00:00.037
Call the function from a SQL query to get employees with total salary >= 12000. Set autotrace on.
SQL> SET AUTOTRACE ON EXPLAIN;
Autotrace Enabled
Displays the execution plan only.
SQL> SELECT * FROM employees where total_sal(salary,commission_pct) > 12000;
EMPLOYEE_ID FIRST_NAME LAST_NAME EMAIL PHONE_NUMBER HIRE_DATE JOB_ID SALARY COMMISSION_PCT MANAGER_ID DEPARTMENT_ID
______________ _____________ ____________ ___________ _____________________ ____________ __________ _________ _________________ _____________ ________________
100 Steven King SKING 515.123.4567 17-JUN-03 AD_PRES 24000 90
101 Neena Kochhar NKOCHHAR 515.123.4568 21-SEP-05 AD_VP 17000 100 90
102 Lex De Haan LDEHAAN 515.123.4569 13-JAN-01 AD_VP 17000 100 90
108 Nancy Greenberg NGREENBE 515.124.4569 17-AUG-02 FI_MGR 12008 101 100
145 John Russell JRUSSEL 011.44.1344.429268 01-OCT-04 SA_MAN 14000 0.4 100 80
146 Karen Partners KPARTNER 011.44.1344.467268 05-JAN-05 SA_MAN 13500 0.3 100 80
147 Alberto Errazuriz AERRAZUR 011.44.1344.429278 10-MAR-05 SA_MAN 12000 0.3 100 80
201 Michael Hartstein MHARTSTE 515.123.5555 17-FEB-04 MK_MAN 13000 100 20
205 Shelley Higgins SHIGGINS 515.123.8080 07-JUN-02 AC_MGR 12008 101 110
90000 Rishin Mitra RMitra 12345 07-JUN-02 IT_PROG 12008 210
99999 Darryl G DG 67890 07-JUN-02 IT_PROG 12008 300
11 rows selected.
PLAN_TABLE_OUTPUT
_________________________________________________________________________
SQL_ID g4c5d1rzy0yda, child number 0
-------------------------------------
SELECT * FROM employees where total_sal(salary,commission_pct) > 12000
Plan hash value: 1445457117
------------------------------------------------
| Id | Operation | Name | E-Rows |
------------------------------------------------
| 0 | SELECT STATEMENT | | |
|* 1 | TABLE ACCESS FULL| EMPLOYEES | 5 |
------------------------------------------------
Predicate Information (identified by operation id):
PLAN_TABLE_OUTPUT
____________________________________________________________________________________
---------------------------------------------------
1 - filter("TOTAL_SAL"("SALARY","COMMISSION_PCT")>12000)
Note
-----
- Warning: basic plan statistics not available. These are only collected when:
* hint 'gather_plan_statistics' is used for the statement or
* parameter 'statistics_level' is set to 'ALL', at session or system level
Elapsed: 00:00:00.376
Notice, that in the predicate section, we see the function total_sal.
Now, let’s set the parameter sql_transpiler ON. I will do it at the session level.
SQL> alter session set sql_transpiler = ON;
Session altered.
Elapsed: 00:00:00.003
Flush the shared pool to make sure that you get a hard-parse.
SQL> ALTER SYSTEM FLUSH shared_pool;
System altered.
Elapsed: 00:00:01.063
Now, run the sql statement again.
SQL> SELECT * FROM employees where total_sal(salary,commission_pct) > 12000;
EMPLOYEE_ID FIRST_NAME LAST_NAME EMAIL PHONE_NUMBER HIRE_DATE JOB_ID SALARY COMMISSION_PCT MANAGER_ID DEPARTMENT_ID
______________ _____________ ____________ ___________ _____________________ ____________ __________ _________ _________________ _____________ ________________
100 Steven King SKING 515.123.4567 17-JUN-03 AD_PRES 24000 90
101 Neena Kochhar NKOCHHAR 515.123.4568 21-SEP-05 AD_VP 17000 100 90
102 Lex De Haan LDEHAAN 515.123.4569 13-JAN-01 AD_VP 17000 100 90
108 Nancy Greenberg NGREENBE 515.124.4569 17-AUG-02 FI_MGR 12008 101 100
145 John Russell JRUSSEL 011.44.1344.429268 01-OCT-04 SA_MAN 14000 0.4 100 80
146 Karen Partners KPARTNER 011.44.1344.467268 05-JAN-05 SA_MAN 13500 0.3 100 80
147 Alberto Errazuriz AERRAZUR 011.44.1344.429278 10-MAR-05 SA_MAN 12000 0.3 100 80
201 Michael Hartstein MHARTSTE 515.123.5555 17-FEB-04 MK_MAN 13000 100 20
205 Shelley Higgins SHIGGINS 515.123.8080 07-JUN-02 AC_MGR 12008 101 110
90000 Rishin Mitra RMitra 12345 07-JUN-02 IT_PROG 12008 210
99999 Darryl G DG 67890 07-JUN-02 IT_PROG 12008 300
11 rows selected.
PLAN_TABLE_OUTPUT
_________________________________________________________________________
SQL_ID g4c5d1rzy0yda, child number 0
-------------------------------------
SELECT * FROM employees where total_sal(salary,commission_pct) > 12000
Plan hash value: 1445457117
------------------------------------------------
| Id | Operation | Name | E-Rows |
------------------------------------------------
| 0 | SELECT STATEMENT | | |
|* 1 | TABLE ACCESS FULL| EMPLOYEES | 5 |
------------------------------------------------
Predicate Information (identified by operation id):
PLAN_TABLE_OUTPUT
____________________________________________________________________________________
---------------------------------------------------
1 - filter("SALARY"+NVL("COMMISSION_PCT",0)*"SALARY"/100>12000)
Note
-----
- Warning: basic plan statistics not available. These are only collected when:
* hint 'gather_plan_statistics' is used for the statement or
* parameter 'statistics_level' is set to 'ALL', at session or system level
Elapsed: 00:00:00.494
Notice the change here in the predicate section. Now, you see that the function has now been transformed into the underlying SQL expression. That is automatic transpiler in action.
Now let us look at if it has any impact on the performance or execution times at all.
Performance Test
I will set the transpiler off and flush the shared pool.
SQL> ALTER SYSTEM FLUSH shared_pool;
System altered.
Elapsed: 00:00:00.310
SQL> alter session set sql_transpiler = OFF;
Session altered.
Elapsed: 00:00:00.004
We write a PL/SQL block and loop around 100000 times and call the function total_sal. Let us see how long it takes.
SQL> VAR r
SQL> BEGIN
FOR i IN 1..100000 LOOP
SELECT count(1) INTO :r FROM employees WHERE total_sal(salary,commission_pct) > 12000;
END LOOP;
END;
/
PL/SQL procedure successfully completed.
Elapsed: 00:00:30.068
Notice that it took around 30 seconds to execute the PL/SQL block.
Now I will flush the shared pool and set the transpiler ON.
SQL> ALTER SYSTEM FLUSH shared_pool;
System altered.
Elapsed: 00:00:00.301
SQL> alter session set sql_transpiler = ON;
Session altered.
Elapsed: 00:00:00.003
We execute the PL/SQL block once again, but this time with transpiler switched on.
SQL> VAR r NUMBER
SQL> BEGIN
FOR i IN 1..100000 LOOP
SELECT count(1) INTO :r FROM employees WHERE total_sal(salary,commission_pct) > 12000;
END LOOP;
END;
/
PL/SQL procedure successfully completed.
Elapsed: 00:00:05.805
Oh wow, this time the same PL/SQL block took almost 6 seconds. The performance of the block improved significantly. To ensure it was not a fluke, let me execute the PL/SQL block once again.
SQL> BEGIN
FOR i IN 1..100000 LOOP
SELECT count(1) INTO :r FROM employees WHERE total_sal(salary,commission_pct) > 12000;
END LOOP;
END;
/
PL/SQL procedure successfully completed.
Elapsed: 00:00:03.400
This time it took even less time.
Hence, we have proved setting the transpiler ON could significantly reduce the context switches, which has significantly reduced the execution time.
However, there are certain limitations for a transpiler. Details of the eligibility/ineligibility of the PL/SQL constructs can be found in this Oracle note.
Happy transpiling.
References
RishOraDev's Oracle Blogs 