Switch from preallocating all databases to lazy allocation,

reducing overhead and improving scalability for large database counts.
valkey-io · Jan 27, 2025 · 33da0ea · 33da0ea
1 parent 3f21705
commit 33da0ea
Show file tree

Hide file tree

Showing 13 changed files with 165 additions and 105 deletions.
diff --git a/src/aof.c b/src/aof.c
@@ -2210,8 +2210,8 @@ int rewriteAppendOnlyFileRio(rio *aof) {
 
     for (j = 0; j < server.dbnum; j++) {
         char selectcmd[] = "*2\r\n$6\r\nSELECT\r\n";
-        serverDb *db = server.db + j;
-        if (kvstoreSize(db->keys) == 0) continue;
+        if (databaseEmpty(j)) continue;
+        serverDb *db = server.db[j];
 
         /* SELECT the new DB */
         if (rioWrite(aof, selectcmd, sizeof(selectcmd) - 1) == 0) goto werr;

diff --git a/src/cluster.c b/src/cluster.c
@@ -812,7 +812,7 @@ static int shouldReturnTlsInfo(void) {
 }
 
 unsigned int countKeysInSlot(unsigned int slot) {
-    return kvstoreHashtableSize(server.db->keys, slot);
+    return kvstoreHashtableSize(server.db[0]->keys, slot);
 }
 
 void clusterCommandHelp(client *c) {
@@ -910,7 +910,7 @@ void clusterCommand(client *c) {
         unsigned int numkeys = maxkeys > keys_in_slot ? keys_in_slot : maxkeys;
         addReplyArrayLen(c, numkeys);
         kvstoreHashtableIterator *kvs_di = NULL;
-        kvs_di = kvstoreGetHashtableIterator(server.db->keys, slot, 0);
+        kvs_di = kvstoreGetHashtableIterator(server.db[0]->keys, slot, 0);
         for (unsigned int i = 0; i < numkeys; i++) {
             void *next;
             serverAssert(kvstoreHashtableIteratorNext(kvs_di, &next));
@@ -1099,7 +1099,7 @@ getNodeByQuery(client *c, struct serverCommand *cmd, robj **argv, int argc, int
              * NODE <node-id>. */
             int flags = LOOKUP_NOTOUCH | LOOKUP_NOSTATS | LOOKUP_NONOTIFY | LOOKUP_NOEXPIRE;
             if ((migrating_slot || importing_slot) && !pubsubshard_included) {
-                if (lookupKeyReadWithFlags(&server.db[0], thiskey, flags) == NULL)
+                if (lookupKeyReadWithFlags(server.db[0], thiskey, flags) == NULL)
                     missing_keys++;
                 else
                     existing_keys++;

diff --git a/src/cluster_legacy.c b/src/cluster_legacy.c
@@ -5711,7 +5711,7 @@ int verifyClusterConfigWithData(void) {
 
     /* Make sure we only have keys in DB0. */
     for (j = 1; j < server.dbnum; j++) {
-        if (kvstoreSize(server.db[j].keys)) return C_ERR;
+        if (!databaseEmpty(j)) return C_ERR;
     }
 
     /* Check that all the slots we see populated memory have a corresponding
@@ -6347,19 +6347,20 @@ unsigned int delKeysInSlot(unsigned int hashslot) {
 
     kvstoreHashtableIterator *kvs_di = NULL;
     void *next;
-    kvs_di = kvstoreGetHashtableIterator(server.db->keys, hashslot, HASHTABLE_ITER_SAFE);
+    serverDb *db = server.db[0];
+    kvs_di = kvstoreGetHashtableIterator(db->keys, hashslot, HASHTABLE_ITER_SAFE);
     while (kvstoreHashtableIteratorNext(kvs_di, &next)) {
         robj *valkey = next;
         enterExecutionUnit(1, 0);
         sds sdskey = objectGetKey(valkey);
         robj *key = createStringObject(sdskey, sdslen(sdskey));
-        dbDelete(&server.db[0], key);
-        propagateDeletion(&server.db[0], key, server.lazyfree_lazy_server_del);
-        signalModifiedKey(NULL, &server.db[0], key);
+        dbDelete(db, key);
+        propagateDeletion(db, key, server.lazyfree_lazy_server_del);
+        signalModifiedKey(NULL, db, key);
         /* The keys are not actually logically deleted from the database, just moved to another node.
          * The modules needs to know that these keys are no longer available locally, so just send the
          * keyspace notification to the modules, but not to clients. */
-        moduleNotifyKeyspaceEvent(NOTIFY_GENERIC, "del", key, server.db[0].id);
+        moduleNotifyKeyspaceEvent(NOTIFY_GENERIC, "del", key, db->id);
         exitExecutionUnit();
         postExecutionUnitOperations();
         decrRefCount(key);
@@ -6827,7 +6828,7 @@ int clusterCommandSpecial(client *c) {
         }
     } else if (!strcasecmp(c->argv[1]->ptr, "flushslots") && c->argc == 2) {
         /* CLUSTER FLUSHSLOTS */
-        if (kvstoreSize(server.db[0].keys) != 0) {
+        if (!databaseEmpty(0)) {
             addReplyError(c, "DB must be empty to perform CLUSTER FLUSHSLOTS.");
             return 1;
         }
@@ -6968,7 +6969,7 @@ int clusterCommandSpecial(client *c) {
         /* If the instance is currently a primary, it should have no assigned
          * slots nor keys to accept to replicate some other node.
          * Replicas can switch to another primary without issues. */
-        if (clusterNodeIsPrimary(myself) && (myself->numslots != 0 || kvstoreSize(server.db[0].keys) != 0)) {
+        if (clusterNodeIsPrimary(myself) && (myself->numslots != 0 || !databaseEmpty(0))) {
             addReplyError(c, "To set a master the node must be empty and "
                              "without assigned slots.");
             return 1;

diff --git a/src/db.c b/src/db.c
@@ -571,7 +571,7 @@ robj *dbUnshareStringValue(serverDb *db, robj *key, robj *o) {
  * The dbnum can be -1 if all the DBs should be emptied, or the specified
  * DB index if we want to empty only a single database.
  * The function returns the number of keys removed from the database(s). */
-long long emptyDbStructure(serverDb *dbarray, int dbnum, int async, void(callback)(hashtable *)) {
+long long emptyDbStructure(serverDb **dbarray, int dbnum, int async, void(callback)(hashtable *)) {
     long long removed = 0;
     int startdb, enddb;
 
@@ -583,16 +583,17 @@ long long emptyDbStructure(serverDb *dbarray, int dbnum, int async, void(callbac
     }
 
     for (int j = startdb; j <= enddb; j++) {
-        removed += kvstoreSize(dbarray[j].keys);
+        if (dbarray[j] == NULL) continue;
+        removed += kvstoreSize(dbarray[j]->keys);
         if (async) {
-            emptyDbAsync(&dbarray[j]);
+            emptyDbAsync(dbarray[j]);
         } else {
-            kvstoreEmpty(dbarray[j].keys, callback);
-            kvstoreEmpty(dbarray[j].expires, callback);
+            kvstoreEmpty(dbarray[j]->keys, callback);
+            kvstoreEmpty(dbarray[j]->expires, callback);
         }
         /* Because all keys of database are removed, reset average ttl. */
-        dbarray[j].avg_ttl = 0;
-        dbarray[j].expires_cursor = 0;
+        dbarray[j]->avg_ttl = 0;
+        dbarray[j]->expires_cursor = 0;
     }
 
     return removed;
@@ -669,28 +670,31 @@ serverDb *initTempDb(void) {
 }
 
 /* Discard tempDb, it's always async. */
-void discardTempDb(serverDb *tempDb) {
+void discardTempDb(serverDb **tempDb) {
     /* Release temp DBs. */
     emptyDbStructure(tempDb, -1, 1, NULL);
     for (int i = 0; i < server.dbnum; i++) {
-        kvstoreRelease(tempDb[i].keys);
-        kvstoreRelease(tempDb[i].expires);
+        if (tempDb[i]) {
+            kvstoreRelease(tempDb[i]->keys);
+            kvstoreRelease(tempDb[i]->expires);
+        }
     }
-
     zfree(tempDb);
 }
 
 int selectDb(client *c, int id) {
     if (id < 0 || id >= server.dbnum) return C_ERR;
-    c->db = &server.db[id];
+    initDatabase(id);
+    c->db = server.db[id];
     return C_OK;
 }
 
 long long dbTotalServerKeyCount(void) {
     long long total = 0;
     int j;
     for (j = 0; j < server.dbnum; j++) {
-        total += kvstoreSize(server.db[j].keys);
+        if (databaseEmpty(j)) continue;
+        total += kvstoreSize(server.db[j]->keys);
     }
     return total;
 }
@@ -721,8 +725,9 @@ void signalFlushedDb(int dbid, int async) {
     }
 
     for (int j = startdb; j <= enddb; j++) {
-        scanDatabaseForDeletedKeys(&server.db[j], NULL);
-        touchAllWatchedKeysInDb(&server.db[j], NULL);
+        if (server.db[j] == NULL) continue;
+        scanDatabaseForDeletedKeys(server.db[j], NULL);
+        touchAllWatchedKeysInDb(server.db[j], NULL);
     }
 
     trackingInvalidateKeysOnFlush(async);
@@ -1641,8 +1646,10 @@ void scanDatabaseForDeletedKeys(serverDb *emptied, serverDb *replaced_with) {
 int dbSwapDatabases(int id1, int id2) {
     if (id1 < 0 || id1 >= server.dbnum || id2 < 0 || id2 >= server.dbnum) return C_ERR;
     if (id1 == id2) return C_OK;
-    serverDb aux = server.db[id1];
-    serverDb *db1 = &server.db[id1], *db2 = &server.db[id2];
+    initDatabase(id1);
+    initDatabase(id2);
+    serverDb aux = *server.db[id1];
+    serverDb *db1 = server.db[id1], *db2 = server.db[id2];
 
     /* Swapdb should make transaction fail if there is any
      * client watching keys */
@@ -1683,10 +1690,13 @@ int dbSwapDatabases(int id1, int id2) {
 /* Logically, this discards (flushes) the old main database, and apply the newly loaded
  * database (temp) as the main (active) database, the actual freeing of old database
  * (which will now be placed in the temp one) is done later. */
-void swapMainDbWithTempDb(serverDb *tempDb) {
+void swapMainDbWithTempDb(serverDb **tempDb) {
     for (int i = 0; i < server.dbnum; i++) {
-        serverDb aux = server.db[i];
-        serverDb *activedb = &server.db[i], *newdb = &tempDb[i];
+        if (tempDb[i] == NULL && server.db[i] == NULL) continue;
+        if (tempDb[i] == NULL) tempDb[i] = createDatabase(i);
+        if (server.db[i] == NULL) server.db[i] = createDatabase(i);
+        serverDb aux = *server.db[i];
+        serverDb *activedb = server.db[i], *newdb = tempDb[i];
 
         /* Swapping databases should make transaction fail if there is any
          * client watching keys. */

diff --git a/src/debug.c b/src/debug.c
@@ -289,8 +289,8 @@ void computeDatasetDigest(unsigned char *final) {
     memset(final, 0, 20); /* Start with a clean result */
 
     for (int j = 0; j < server.dbnum; j++) {
-        serverDb *db = server.db + j;
-        if (kvstoreSize(db->keys) == 0) continue;
+        serverDb *db = server.db[j];
+        if (db == NULL || kvstoreSize(db->keys) == 0) continue;
         kvstoreIterator *kvs_it = kvstoreIteratorInit(db->keys, HASHTABLE_ITER_SAFE | HASHTABLE_ITER_PREFETCH_VALUES);
 
         /* hash the DB id, so the same dataset moved in a different DB will lead to a different digest */
@@ -907,14 +907,20 @@ void debugCommand(client *c) {
         if (c->argc >= 4 && !strcasecmp(c->argv[3]->ptr, "full")) full = 1;
 
         stats = sdscatprintf(stats, "[Dictionary HT]\n");
-        kvstoreGetStats(server.db[dbid].keys, buf, sizeof(buf), full);
-        stats = sdscat(stats, buf);
+        serverDb *db = server.db[dbid];
+        if (db) {
+            kvstoreGetStats(db->keys, buf, sizeof(buf), full);
+            stats = sdscat(stats, buf);
+        }
 
         stats = sdscatprintf(stats, "[Expires HT]\n");
-        kvstoreGetStats(server.db[dbid].expires, buf, sizeof(buf), full);
-        stats = sdscat(stats, buf);
+        if (db) {
+            kvstoreGetStats(db->expires, buf, sizeof(buf), full);
+            stats = sdscat(stats, buf);
+        }
 
         addReplyVerbatim(c, stats, sdslen(stats), "txt");
+
         sdsfree(stats);
     } else if (!strcasecmp(c->argv[1]->ptr, "htstats-key") && c->argc >= 3) {
         int full = 0;

diff --git a/src/defrag.c b/src/defrag.c
@@ -726,7 +726,7 @@ static void defragModule(serverDb *db, robj *obj) {
 /* for each key we scan in the main dict, this function will attempt to defrag
  * all the various pointers it has. */
 static void defragKey(defragKeysCtx *ctx, robj **elemref) {
-    serverDb *db = &server.db[ctx->dbid];
+    serverDb *db = server.db[ctx->dbid];
     int slot = ctx->kvstate.slot;
     robj *newob, *ob;
     unsigned char *newzl;
@@ -987,7 +987,7 @@ static doneStatus defragStageKvstoreHelper(monotime endtime,
 static doneStatus defragStageDbKeys(monotime endtime, void *target, void *privdata) {
     UNUSED(privdata);
     int dbid = (uintptr_t)target;
-    serverDb *db = &server.db[dbid];
+    serverDb *db = server.db[dbid];
 
     static defragKeysCtx ctx; // STATIC - this persists
     if (endtime == 0) {
@@ -1005,7 +1005,7 @@ static doneStatus defragStageDbKeys(monotime endtime, void *target, void *privda
 static doneStatus defragStageExpiresKvstore(monotime endtime, void *target, void *privdata) {
     UNUSED(privdata);
     int dbid = (uintptr_t)target;
-    serverDb *db = &server.db[dbid];
+    serverDb *db = server.db[dbid];
     return defragStageKvstoreHelper(endtime, db->expires,
                                     scanHashtableCallbackCountScanned, NULL, NULL);
 }
@@ -1273,6 +1273,7 @@ static void beginDefragCycle(void) {
     defrag.remaining_stages = listCreate();
 
     for (int dbid = 0; dbid < server.dbnum; dbid++) {
+        if (databaseEmpty(dbid)) continue;
         addDefragStage(defragStageDbKeys, (void *)(uintptr_t)dbid, NULL);
         addDefragStage(defragStageExpiresKvstore, (void *)(uintptr_t)dbid, NULL);
     }

diff --git a/src/evict.c b/src/evict.c
@@ -568,7 +568,8 @@ int performEvictions(void) {
                  * so to start populate the eviction pool sampling keys from
                  * every DB. */
                 for (i = 0; i < server.dbnum; i++) {
-                    db = server.db + i;
+                    db = server.db[i];
+                    if (db == NULL) continue;;
                     kvstore *kvs;
                     if (server.maxmemory_policy & MAXMEMORY_FLAG_ALLKEYS) {
                         kvs = db->keys;
@@ -601,9 +602,9 @@ int performEvictions(void) {
 
                     kvstore *kvs;
                     if (server.maxmemory_policy & MAXMEMORY_FLAG_ALLKEYS) {
-                        kvs = server.db[bestdbid].keys;
+                        kvs = server.db[bestdbid]->keys;
                     } else {
-                        kvs = server.db[bestdbid].expires;
+                        kvs = server.db[bestdbid]->expires;
                     }
                     void *entry = NULL;
                     int found = kvstoreHashtableFind(kvs, pool[k].slot, pool[k].key, &entry);
@@ -634,7 +635,8 @@ int performEvictions(void) {
              * incrementally visit all DBs. */
             for (i = 0; i < server.dbnum; i++) {
                 j = (++next_db) % server.dbnum;
-                db = server.db + j;
+                db = server.db[j];
+                if (db == NULL) continue;;
                 kvstore *kvs;
                 if (server.maxmemory_policy == MAXMEMORY_ALLKEYS_RANDOM) {
                     kvs = db->keys;
@@ -653,7 +655,7 @@ int performEvictions(void) {
 
         /* Finally remove the selected key. */
         if (bestkey) {
-            db = server.db + bestdbid;
+            db = server.db[bestdbid];
             robj *keyobj = createStringObject(bestkey, sdslen(bestkey));
             /* We compute the amount of memory freed by db*Delete() alone.
              * It is possible that actually the memory needed to propagate

diff --git a/src/expire.c b/src/expire.c
@@ -234,7 +234,7 @@ void activeExpireCycle(int type) {
         data.ttl_sum = 0;
         data.ttl_samples = 0;
 
-        serverDb *db = server.db + (current_db % server.dbnum);
+        serverDb *db = server.db[(current_db % server.dbnum)];
         data.db = db;
 
         int db_done = 0; /* The scan of the current DB is done? */
@@ -245,13 +245,17 @@ void activeExpireCycle(int type) {
          * distribute the time evenly across DBs. */
         current_db++;
 
-        if (kvstoreSize(db->expires)) dbs_performed++;
+        if (db && kvstoreSize(db->expires)) dbs_performed++;
 
         /* Continue to expire if at the end of the cycle there are still
          * a big percentage of keys to expire, compared to the number of keys
          * we scanned. The percentage, stored in config_cycle_acceptable_stale
          * is not fixed, but depends on the configured "expire effort". */
         do {
+            if (db == NULL) {
+                break; /* DB not allocated since it was never used */
+            }
+
             unsigned long num;
             iteration++;
 
@@ -421,11 +425,11 @@ void expireReplicaKeys(void) {
         int dbid = 0;
         while (dbids && dbid < server.dbnum) {
             if ((dbids & 1) != 0) {
-                serverDb *db = server.db + dbid;
-                robj *expire = dbFindExpires(db, keyname);
+                serverDb *db = server.db[dbid];
+                robj *expire = db == NULL ? NULL : dbFindExpires(db, keyname);
                 int expired = 0;
 
-                if (expire && activeExpireCycleTryExpire(server.db + dbid, expire, start)) {
+                if (expire && activeExpireCycleTryExpire(db, expire, start)) {
                     expired = 1;
                     /* Propagate the DEL (writable replicas do not propagate anything to other replicas,
                      * but they might propagate to AOF) and trigger module hooks. */

diff --git a/src/object.c b/src/object.c
@@ -1357,8 +1357,8 @@ struct serverMemOverhead *getMemoryOverheadData(void) {
     mem_total += mh->functions_caches;
 
     for (j = 0; j < server.dbnum; j++) {
-        serverDb *db = server.db + j;
-        if (!kvstoreNumAllocatedHashtables(db->keys)) continue;
+        serverDb *db = server.db[j];
+        if (db == NULL || !kvstoreNumAllocatedHashtables(db->keys)) continue;
 
         unsigned long long keyscount = kvstoreSize(db->keys);