- Berkeley DB Reference Guide:
- Transaction Protected Applications
|
|
The third reason listed for using transactions was atomicity. Consider an application suite where multiple threads of control (multiple processes or threads in one or more processes) are changing the values associated with a key in one or more databases. Specifically, they are taking the current value, incrementing it, and then storing it back into the database.
Such an application requires atomicity. Since we want to change a value in the database, we must make sure that once we read it, no other thread of control modifies it. For example, assume that both thread #1 and thread #2 are doing similar operations in the database, where thread #1 is incrementing records by 3, and thread #2 is incrementing records by 5. We want to increment the record by a total of 8. If the operations interleave in the right (well, wrong) order, that is not what will happen:
thread #1 read record: the value is 2 thread #2 read record: the value is 2 thread #2 write record + 5 back into the database (new value 7) thread #1 write record + 3 back into the database (new value 5)
As you can see, instead of incrementing the record by a total of 8, we've only incremented it by 3, because thread #1 overwrote thread #2's change. By wrapping the operations in transactions, we ensure that this cannot happen. In a transaction, when the first thread reads the record, locks are acquired that will not be released until the transaction finishes, guaranteeing that all other readers and writers will block, waiting for the first thread's transaction to complete (or to be aborted).
Here is an example function that does transaction-protected increments on database records to ensure atomicity.
int
main(int argc, char *argv)
{
extern char *optarg;
extern int optind;
DB *db_cats, *db_color, *db_fruit;
DB_ENV *dbenv;
pthread_t ptid;
int ch;
while ((ch = getopt(argc, argv, "")) != EOF)
switch (ch) {
case '?':
default:
usage();
}
argc -= optind;
argv += optind;
env_dir_create();
env_open(&dbenv);
/* Open database: Key is fruit class; Data is specific type. */
db_open(dbenv, &db_fruit, "fruit", 0);
/* Open database: Key is a color; Data is an integer. */
db_open(dbenv, &db_color, "color", 0);
/*
* Open database:
* Key is a name; Data is: company name, address, cat breeds.
*/
db_open(dbenv, &db_cats, "cats", 1);
add_fruit(dbenv, db_fruit, "apple", "yellow delicious");
add_color(dbenv, db_color, "blue", 0);
add_color(dbenv, db_color, "blue", 3);
return (0);
}
void
add_color(DB_ENV *dbenv, DB *dbp, char *color, int increment)
{
DBT key, data;
DB_TXN *tid;
int original, ret;
char buf64;
/* Initialization. */
memset(&key, 0, sizeof(key));
key.data = color;
key.size = strlen(color);
memset(&data, 0, sizeof(data));
data.flags = DB_DBT_MALLOC;
for (;;) {
/* Begin the transaction. */
if ((ret = txn_begin(dbenv, NULL, &tid, 0)) != 0) {
dbenv->err(dbenv, ret, "txn_begin");
exit (1);
}
/*
* Get the key. If it exists, we increment the value. If it
* doesn't exist, we create it.
*/
switch (ret = dbp->get(dbp, tid, &key, &data, 0)) {
case 0:
original = atoi(data.data);
break;
case DB_LOCK_DEADLOCK:
/* Deadlock: retry the operation. */
if ((ret = txn_abort(tid)) != 0) {
dbenv->err(dbenv, ret, "txn_abort");
exit (1);
}
continue;
case DB_NOTFOUND:
original = 0;
break;
default:
/* Error: run recovery. */
dbenv->err(
dbenv, ret, "dbc->get: %s/%d", color, increment);
exit (1);
}
if (data.data != NULL)
free(data.data);
/* Create the new data item. */
(void)snprintf(buf, sizeof(buf), "%d", original + increment);
data.data = buf;
data.size = strlen(buf) + 1;
/* Store the new value. */
switch (ret = dbp->put(dbp, tid, &key, &data, 0)) {
case 0:
/* Success: commit the change. */
if ((ret = txn_commit(tid, 0)) != 0) {
dbenv->err(dbenv, ret, "txn_commit");
exit (1);
}
return;
case DB_LOCK_DEADLOCK:
/* Deadlock: retry the operation. */
if ((ret = txn_abort(tid)) != 0) {
dbenv->err(dbenv, ret, "txn_abort");
exit (1);
}
break;
default:
/* Error: run recovery. */
dbenv->err(
dbenv, ret, "dbc->put: %s/%d", color, increment);
exit (1);
}
}
}
Any number of operations, on any number of databases, can be included in a single transaction to ensure atomicity of the operations. There is, however, a trade-off between the number of operations included in a single transaction and both throughput and the possibility of deadlock. The reason for this is because transactions acquire locks throughout their lifetime, and do not release them until transaction commit or abort. So, the more operations included in a transaction, the more likely that a transaction will block other operations and that deadlock will occur. However, each transaction commit requires a synchronous disk I/O, so grouping multiple operations into a transaction can increase overall throughput. (There is one exception to this. The DB_TXN_NOSYNC option causes transactions to exhibit the ACI (atomicity, consistency and isolation) properties, but not D (durability), avoiding the synchronous disk I/O on transaction commit and greatly increasing transaction throughput for some applications.
When applications do create complex transactions, they often avoid having more than one complex transaction at a time, as simple operations like a single DB->put are unlikely to deadlock with each other or the complex transaction, while multiple complex transactions are likely to deadlock with each other as they will both acquire many locks over their lifetime. Alternatively, complex transactions can be broken up into smaller sets of operations, and each of those sets may be encapsulated in a nested transaction. Because nested transactions may be individually aborted and retried without causing the entire transaction to be aborted, this allows complex transactions to proceed even in the face of heavy contention, repeatedly trying the sub-operations until they succeed.
It is also helpful to order operations within a transaction, that is, access the databases and items within the databases in the same order, to the extent possible, in all transactions. Accessing databases and items in different orders greatly increases the likelihood of operations being blocked and failing due to deadlocks.
|
|