Improve INSERT-per-second performance of SQLite?

Question

Optimizing SQLite is tricky. Bulk-insert performance of a C application can vary from 85 inserts per second to over 96,000 inserts per second!

Background: We are using SQLite as part of a desktop application. We have large amounts of configuration data stored in XML files that are parsed and loaded into an SQLite database for further processing when the application is initialized. SQLite is ideal for this situation because it's fast, it requires no specialized configuration, and the database is stored on disk as a single file.

Rationale: Initially I was disappointed with the performance I was seeing. It turns-out that the performance of SQLite can vary significantly (both for bulk-inserts and selects) depending on how the database is configured and how you're using the API. It was not a trivial matter to figure out what all of the options and techniques were, so I thought it prudent to create this community wiki entry to share the results with Stack Overflow readers in order to save others the trouble of the same investigations.

The Experiment: Rather than simply talking about performance tips in the general sense (i.e. "Use a transaction!"), I thought it best to write some C code and actually measure the impact of various options. We're going to start with some simple data:

• A 28 MB TAB-delimited text file (approximately 865,000 records) of the complete transit schedule for the city of Toronto
• My test machine is a 3.60 GHz P4 running Windows XP.
• The code is compiled with Visual C++ 2005 as "Release" with "Full Optimization" (/Ox) and Favor Fast Code (/Ot).
• I'm using the SQLite "Amalgamation", compiled directly into my test application. The SQLite version I happen to have is a bit older (3.6.7), but I suspect these results will be comparable to the latest release (please leave a comment if you think otherwise).

Let's write some code!

The Code: A simple C program that reads the text file line-by-line, splits the string into values and then will inserts the data into an SQLite database. In this "baseline" version of the code, the database is created, but we won't actually insert data:

/*************************************************************
    Baseline code to experiment with SQLite performance.

    Input data is a 28 MB TAB-delimited text file of the
    complete Toronto Transit System schedule/route info
    from http://www.toronto.ca/open/datasets/ttc-routes/

**************************************************************/
#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#include <string.h>
#include "sqlite3.h"

#define INPUTDATA "C:\\TTC_schedule_scheduleitem_10-27-2009.txt"
#define DATABASE "c:\\TTC_schedule_scheduleitem_10-27-2009.sqlite"
#define TABLE "CREATE TABLE IF NOT EXISTS TTC (id INTEGER PRIMARY KEY, Route_ID TEXT, Branch_Code TEXT, Version INTEGER, Stop INTEGER, Vehicle_Index INTEGER, Day Integer, Time TEXT)"
#define BUFFER_SIZE 256

int main(int argc, char **argv) {

    sqlite3 * db;
    sqlite3_stmt * stmt;
    char * sErrMsg = 0;
    char * tail = 0;
    int nRetCode;
    int n = 0;

    clock_t cStartClock;

    FILE * pFile;
    char sInputBuf [BUFFER_SIZE] = "\0";

    char * sRT = 0;  /* Route */
    char * sBR = 0;  /* Branch */
    char * sVR = 0;  /* Version */
    char * sST = 0;  /* Stop Number */
    char * sVI = 0;  /* Vehicle */
    char * sDT = 0;  /* Date */
    char * sTM = 0;  /* Time */

    char sSQL [BUFFER_SIZE] = "\0";

    /*********************************************/
    /* Open the Database and create the Schema */
    sqlite3_open(DATABASE, &db);
    sqlite3_exec(db, TABLE, NULL, NULL, &sErrMsg);

    /*********************************************/
    /* Open input file and import into Database*/
    cStartClock = clock();

    pFile = fopen (INPUTDATA,"r");
    while (!feof(pFile)) {

        fgets (sInputBuf, BUFFER_SIZE, pFile);

        sRT = strtok (sInputBuf, "\t");     /* Get Route */
        sBR = strtok (NULL, "\t");            /* Get Branch */
        sVR = strtok (NULL, "\t");            /* Get Version */
        sST = strtok (NULL, "\t");            /* Get Stop Number */
        sVI = strtok (NULL, "\t");            /* Get Vehicle */
        sDT = strtok (NULL, "\t");            /* Get Date */
        sTM = strtok (NULL, "\t");            /* Get Time */

        /* ACTUAL INSERT WILL GO HERE */

        n++;
    }
    fclose (pFile);

    printf("Imported %d records in %4.2f seconds\n", n, (clock() - cStartClock) / (double)CLOCKS_PER_SEC);

    sqlite3_close(db);
    return 0;
}

---

The "Control"

Running the code as-is doesn't actually perform any database operations, but it will give us an idea of how fast the raw C file I/O and string processing operations are.

Imported 864913 records in 0.94 seconds

Great! We can do 920,000 inserts per second, provided we don't actually do any inserts :-)

---

The "Worst-Case-Scenario"

We're going to generate the SQL string using the values read from the file and invoke that SQL operation using sqlite3_exec:

sprintf(sSQL, "INSERT INTO TTC VALUES (NULL, '%s', '%s', '%s', '%s', '%s', '%s', '%s')", sRT, sBR, sVR, sST, sVI, sDT, sTM);
sqlite3_exec(db, sSQL, NULL, NULL, &sErrMsg);

This is going to be slow because the SQL will be compiled into VDBE code for every insert and every insert will happen in its own transaction. How slow?

Imported 864913 records in 9933.61 seconds

Yikes! 2 hours and 45 minutes! That's only 85 inserts per second.

Using a Transaction

By default, SQLite will evaluate every INSERT / UPDATE statement within a unique transaction. If performing a large number of inserts, it's advisable to wrap your operation in a transaction:

sqlite3_exec(db, "BEGIN TRANSACTION", NULL, NULL, &sErrMsg);

pFile = fopen (INPUTDATA,"r");
while (!feof(pFile)) {

    ...

}
fclose (pFile);

sqlite3_exec(db, "END TRANSACTION", NULL, NULL, &sErrMsg);

Imported 864913 records in 38.03 seconds

That's better. Simply wrapping all of our inserts in a single transaction improved our performance to 23,000 inserts per second.

Using a Prepared Statement

Using a transaction was a huge improvement, but recompiling the SQL statement for every insert doesn't make sense if we using the same SQL over-and-over. Let's use sqlite3_prepare_v2 to compile our SQL statement once and then bind our parameters to that statement using sqlite3_bind_text:

/* Open input file and import into the database */
cStartClock = clock();

sprintf(sSQL, "INSERT INTO TTC VALUES (NULL, @RT, @BR, @VR, @ST, @VI, @DT, @TM)");
sqlite3_prepare_v2(db,  sSQL, BUFFER_SIZE, &stmt, &tail);

sqlite3_exec(db, "BEGIN TRANSACTION", NULL, NULL, &sErrMsg);

pFile = fopen (INPUTDATA,"r");
while (!feof(pFile)) {

    fgets (sInputBuf, BUFFER_SIZE, pFile);

    sRT = strtok (sInputBuf, "\t");   /* Get Route */
    sBR = strtok (NULL, "\t");        /* Get Branch */
    sVR = strtok (NULL, "\t");        /* Get Version */
    sST = strtok (NULL, "\t");        /* Get Stop Number */
    sVI = strtok (NULL, "\t");        /* Get Vehicle */
    sDT = strtok (NULL, "\t");        /* Get Date */
    sTM = strtok (NULL, "\t");        /* Get Time */

    sqlite3_bind_text(stmt, 1, sRT, -1, SQLITE_TRANSIENT);
    sqlite3_bind_text(stmt, 2, sBR, -1, SQLITE_TRANSIENT);
    sqlite3_bind_text(stmt, 3, sVR, -1, SQLITE_TRANSIENT);
    sqlite3_bind_text(stmt, 4, sST, -1, SQLITE_TRANSIENT);
    sqlite3_bind_text(stmt, 5, sVI, -1, SQLITE_TRANSIENT);
    sqlite3_bind_text(stmt, 6, sDT, -1, SQLITE_TRANSIENT);
    sqlite3_bind_text(stmt, 7, sTM, -1, SQLITE_TRANSIENT);

    sqlite3_step(stmt);

    sqlite3_clear_bindings(stmt);
    sqlite3_reset(stmt);

    n++;
}
fclose (pFile);

sqlite3_exec(db, "END TRANSACTION", NULL, NULL, &sErrMsg);

printf("Imported %d records in %4.2f seconds\n", n, (clock() - cStartClock) / (double)CLOCKS_PER_SEC);

sqlite3_finalize(stmt);
sqlite3_close(db);

return 0;

Imported 864913 records in 16.27 seconds

Nice! There's a little bit more code (don't forget to call sqlite3_clear_bindings and sqlite3_reset), but we've more than doubled our performance to 53,000 inserts per second.

PRAGMA synchronous = OFF

By default, SQLite will pause after issuing a OS-level write command. This guarantees that the data is written to the disk. By setting synchronous = OFF, we are instructing SQLite to simply hand-off the data to the OS for writing and then continue. There's a chance that the database file may become corrupted if the computer suffers a catastrophic crash (or power failure) before the data is written to the platter:

/* Open the database and create the schema */
sqlite3_open(DATABASE, &db);
sqlite3_exec(db, TABLE, NULL, NULL, &sErrMsg);
sqlite3_exec(db, "PRAGMA synchronous = OFF", NULL, NULL, &sErrMsg);

Imported 864913 records in 12.41 seconds

The improvements are now smaller, but we're up to 69,600 inserts per second.

PRAGMA journal_mode = MEMORY

Consider storing the rollback journal in memory by evaluating PRAGMA journal_mode = MEMORY. Your transaction will be faster, but if you lose power or your program crashes during a transaction you database could be left in a corrupt state with a partially-completed transaction:

/* Open the database and create the schema */
sqlite3_open(DATABASE, &db);
sqlite3_exec(db, TABLE, NULL, NULL, &sErrMsg);
sqlite3_exec(db, "PRAGMA journal_mode = MEMORY", NULL, NULL, &sErrMsg);

Imported 864913 records in 13.50 seconds

A little slower than the previous optimization at 64,000 inserts per second.

PRAGMA synchronous = OFF and PRAGMA journal_mode = MEMORY

Let's combine the previous two optimizations. It's a little more risky (in case of a crash), but we're just importing data (not running a bank):

/* Open the database and create the schema */
sqlite3_open(DATABASE, &db);
sqlite3_exec(db, TABLE, NULL, NULL, &sErrMsg);
sqlite3_exec(db, "PRAGMA synchronous = OFF", NULL, NULL, &sErrMsg);
sqlite3_exec(db, "PRAGMA journal_mode = MEMORY", NULL, NULL, &sErrMsg);

Imported 864913 records in 12.00 seconds

Fantastic! We're able to do 72,000 inserts per second.

Using an In-Memory Database

Just for kicks, let's build upon all of the previous optimizations and redefine the database filename so we're working entirely in RAM:

#define DATABASE ":memory:"

Imported 864913 records in 10.94 seconds

It's not super-practical to store our database in RAM, but it's impressive that we can perform 79,000 inserts per second.

Refactoring C Code

Although not specifically an SQLite improvement, I don't like the extra char* assignment operations in the while loop. Let's quickly refactor that code to pass the output of strtok()directly into sqlite3_bind_text(), and let the compiler try to speed things up for us:

pFile = fopen (INPUTDATA,"r");
while (!feof(pFile)) {

    fgets (sInputBuf, BUFFER_SIZE, pFile);

    sqlite3_bind_text(stmt, 1, strtok (sInputBuf, "\t"), -1, SQLITE_TRANSIENT); /* Get Route */
    sqlite3_bind_text(stmt, 2, strtok (NULL, "\t"), -1, SQLITE_TRANSIENT);    /* Get Branch */
    sqlite3_bind_text(stmt, 3, strtok (NULL, "\t"), -1, SQLITE_TRANSIENT);    /* Get Version */
    sqlite3_bind_text(stmt, 4, strtok (NULL, "\t"), -1, SQLITE_TRANSIENT);    /* Get Stop Number */
    sqlite3_bind_text(stmt, 5, strtok (NULL, "\t"), -1, SQLITE_TRANSIENT);    /* Get Vehicle */
    sqlite3_bind_text(stmt, 6, strtok (NULL, "\t"), -1, SQLITE_TRANSIENT);    /* Get Date */
    sqlite3_bind_text(stmt, 7, strtok (NULL, "\t"), -1, SQLITE_TRANSIENT);    /* Get Time */

    sqlite3_step(stmt);        /* Execute the SQL Statement */
    sqlite3_clear_bindings(stmt);    /* Clear bindings */
    sqlite3_reset(stmt);        /* Reset VDBE */

    n++;
}
fclose (pFile);

Note: We are back to using a real database file. In-memory databases are fast, but not necessarily practical

Imported 864913 records in 8.94 seconds

A slight refactoring to the string processing code used in our parameter binding has allowed us to perform 96,700 inserts per second. I think it's safe to say that this is plenty fast. As we start to tweak other variables (i.e. page size, index creation, etc.) this will be our benchmark.

---

Summary (so far)

I hope you're still with me! The reason we started down this road is that bulk-insert performance varies so wildly with SQLite, and it's not always obvious what changes need to be made to speed-up our operation. Using the same compiler (and compiler options), the same version of SQLite and the same data we've optimized our code and our usage of SQLite to go from a worst-case scenario of 85 inserts per second to over 96,000 inserts per second!

---

CREATE INDEX then INSERT vs. INSERT then CREATE INDEX

Before we start measuring SELECT performance, we know that we'll be creating indices. It's been suggested in one of the answers below that when doing bulk inserts, it is faster to create the index after the data has been inserted (as opposed to creating the index first then inserting the data). Let's try:

Create Index then Insert Data

sqlite3_exec(db, "CREATE  INDEX 'TTC_Stop_Index' ON 'TTC' ('Stop')", NULL, NULL, &sErrMsg);
sqlite3_exec(db, "BEGIN TRANSACTION", NULL, NULL, &sErrMsg);
...

Imported 864913 records in 18.13 seconds

Insert Data then Create Index

...
sqlite3_exec(db, "END TRANSACTION", NULL, NULL, &sErrMsg);
sqlite3_exec(db, "CREATE  INDEX 'TTC_Stop_Index' ON 'TTC' ('Stop')", NULL, NULL, &sErrMsg);

Imported 864913 records in 13.66 seconds

As expected, bulk-inserts are slower if one column is indexed, but it does make a difference if the index is created after the data is inserted. Our no-index baseline is 96,000 inserts per second. Creating the index first then inserting data gives us 47,700 inserts per second, whereas inserting the data first then creating the index gives us 63,300 inserts per second.

---

I'd gladly take suggestions for other scenarios to try... And will be compiling similar data for SELECT queries soon.

answer

Several tips:

1. Put inserts/updates in a transaction.
2. For older versions of SQLite - Consider a less paranoid journal mode (pragma journal_mode). There is NORMAL, and then there is OFF, which can significantly increase insert speed if you're not too worried about the database possibly getting corrupted if the OS crashes. If your application crashes the data should be fine. Note that in newer versions, the OFF/MEMORYsettings are not safe for application level crashes.
3. Playing with page sizes makes a difference as well (PRAGMA page_size). Having larger page sizes can make reads and writes go a bit faster as larger pages are held in memory. Note that more memory will be used for your database.
4. If you have indices, consider calling CREATE INDEX after doing all your inserts. This is significantly faster than creating the index and then doing your inserts.
5. You have to be quite careful if you have concurrent access to SQLite, as the whole database is locked when writes are done, and although multiple readers are possible, writes will be locked out. This has been improved somewhat with the addition of a WAL in newer SQLite versions.
6. Take advantage of saving space...smaller databases go faster. For instance, if you have key value pairs, try making the key an INTEGER PRIMARY KEY if possible, which will replace the implied unique row number column in the table.
7. If you are using multiple threads, you can try using the shared page cache, which will allow loaded pages to be shared between threads, which can avoid expensive I/O calls.
8. Don't use !feof(file)!

I've also asked similar questions here and here.

answer

Try using SQLITE_STATIC instead of SQLITE_TRANSIENT for those inserts.

SQLITE_TRANSIENT will cause SQLite to copy the string data before returning.

SQLITE_STATIC tells it that the memory address you gave it will be valid until the query has been performed (which in this loop is always the case). This will save you several allocate, copy and deallocate operations per loop. Possibly a large improvement.

answer

Avoid sqlite3_clear_bindings(stmt);

The code in the test sets the bindings every time through which should be enough.

The C API intro from the SQLite docs says

Prior to calling sqlite3_step() for the first time or immediately after sqlite3_reset(), the application can invoke one of the sqlite3_bind() interfaces to attach values to the parameters. Each call to sqlite3_bind() overrides prior bindings on the same parameter

(see: sqlite.org/cintro.html). There is nothing in the docs for that function saying you must call it in addition to simply setting the bindings.

More detail: Avoid_sqlite3_clear_bindings()

answer

On bulk inserts

Inspired by this post and by the Stack Overflow question that led me here -- Is it possible to insert multiple rows at a time in an SQLite database? -- I've posted my first Git repository:

https://github.com/rdpoor/CreateOrUpdate

which bulk loads an array of ActiveRecords into MySQL, SQLite or PostgreSQL databases. It includes an option to ignore existing records, overwrite them or raise an error. My rudimentary benchmarks show a 10x speed improvement compared to sequential writes -- YMMV.

I'm using it in production code where I frequently need to import large datasets, and I'm pretty happy with it.

answer

If you care only about reading, somewhat faster (but might read stale data) version is to read from multiple connections from multiple threads (connection per-thread).

First find the items, in the table:

SELECT COUNT(*) FROM table

then read in pages (LIMIT/OFFSET):

SELECT * FROM table ORDER BY _ROWID_ LIMIT <limit> OFFSET <offset>

where and are calculated per-thread, like this:

int limit = (count + n_threads - 1)/n_threads;

for each thread:

int offset = thread_index * limit

For our small (200mb) db this made 50-75% speed-up (3.8.0.2 64-bit on Windows 7). Our tables are heavily non-normalized (1000-1500 columns, roughly 100,000 or more rows).

Too many or too little threads won't do it, you need to benchmark and profile yourself.

Also for us, SHAREDCACHE made the performance slower, so I manually put PRIVATECACHE (cause it was enabled globally for us)