diff --git a/src/dnode/CMakeLists.txt b/src/dnode/CMakeLists.txt
index 5735e1a8c18677f7323db6942d115fca681ab105..af2dc2d777bb6d294beecf3b3fa2210ab7230571 100644
--- a/src/dnode/CMakeLists.txt
+++ b/src/dnode/CMakeLists.txt
@@ -27,7 +27,7 @@ IF ((TD_LINUX_64) OR (TD_LINUX_32 AND TD_ARM))
   ENDIF ()
 
   IF (TD_SYNC)
-    TARGET_LINK_LIBRARIES(taosd replica sync)
+    TARGET_LINK_LIBRARIES(taosd balance sync)
   ENDIF ()
 
   SET(PREPARE_ENV_CMD "prepare_env_cmd")
diff --git a/src/dnode/src/dnodeMClient.c b/src/dnode/src/dnodeMClient.c
index 90a093560f7860a3414f803c7549f50f4388797f..38be318c2570a89db07ef86225ffbcec409f3cbc 100644
--- a/src/dnode/src/dnodeMClient.c
+++ b/src/dnode/src/dnodeMClient.c
@@ -23,12 +23,13 @@
 #include "tsync.h"
 #include "ttime.h"
 #include "ttimer.h"
-#include "treplica.h"
+#include "tbalance.h"
+#include "vnode.h"
+#include "mnode.h"
 #include "dnode.h"
 #include "dnodeMClient.h"
 #include "dnodeModule.h"
 #include "dnodeMgmt.h"
-#include "vnode.h"
 
 #define MPEER_CONTENT_LEN 2000
 
@@ -181,7 +182,7 @@ static void dnodeProcessStatusRsp(SRpcMsg *pMsg) {
              tsMnodeInfos.nodeInfos[i].nodeName);
     }
     dnodeSaveMnodeIpList();
-    replicaNotify();
+    sdbUpdateSync();
   }
 
   taosTmrReset(dnodeSendStatusMsg, tsStatusInterval * 1000, NULL, tsDnodeTmr, &tsStatusTimer);
diff --git a/src/inc/mnode.h b/src/inc/mnode.h
index e7ad88d6b66398e14411bcbf9e74ffeedb18cbda..35f7650c20e344f42ba8821fc0abeba50d035fe1 100644
--- a/src/inc/mnode.h
+++ b/src/inc/mnode.h
@@ -24,6 +24,7 @@ int32_t mgmtInitSystem();
 int32_t mgmtStartSystem();
 void    mgmtCleanUpSystem();
 void    mgmtStopSystem();
+void    sdbUpdateSync();
 
 #ifdef __cplusplus
 }
diff --git a/src/inc/treplica.h b/src/inc/tbalance.h
similarity index 71%
rename from src/inc/treplica.h
rename to src/inc/tbalance.h
index 3abed1c4aa258bca9e538b561d3da5599ca3e48e..9ffa6332c60957b2e86f1ad6fec4c64c4a64f43f 100644
--- a/src/inc/treplica.h
+++ b/src/inc/tbalance.h
@@ -13,8 +13,8 @@
  * along with this program. If not, see <http://www.gnu.org/licenses/>.
  */
 
-#ifndef TDENGINE_REPLICA_H
-#define TDENGINE_REPLICA_H
+#ifndef TDENGINE_BALANCE_H
+#define TDENGINE_BALANCE_H
 
 #ifdef __cplusplus
 extern "C" {
@@ -23,13 +23,12 @@ extern "C" {
 struct SVgObj;
 struct SDnodeObj;
 
-int32_t replicaInit();
-void    replicaCleanUp();
-void    replicaNotify();
-void    replicaReset();
-int32_t replicaAllocVnodes(struct SVgObj *pVgroup);
-int32_t replicaForwardReqToPeer(void *pHead);
-int32_t replicaDropDnode(struct SDnodeObj *pDnode);
+int32_t balanceInit();
+void    balanceCleanUp();
+void    balanceNotify();
+void    balanceReset();
+int32_t balanceAllocVnodes(struct SVgObj *pVgroup);
+int32_t balanceDropDnode(struct SDnodeObj *pDnode);
 
 #ifdef __cplusplus
 }
diff --git a/src/mnode/inc/mgmtMnode.h b/src/mnode/inc/mgmtMnode.h
index 1faa616cebe12eddb000db0e97f731c33f7463d4..34b5d19cf21108aa7f9808bed4329fed02ebb9b5 100644
--- a/src/mnode/inc/mgmtMnode.h
+++ b/src/mnode/inc/mgmtMnode.h
@@ -39,10 +39,9 @@ int32_t mgmtGetMnodesNum();
 void *  mgmtGetNextMnode(void *pNode, struct SMnodeObj **pMnode);
 void    mgmtReleaseMnode(struct SMnodeObj *pMnode);
 
-bool    mgmtIsMaster();
-
+char *  mgmtGetMnodeRoleStr();
 void    mgmtGetMnodeIpList(SRpcIpSet *ipSet, bool usePublicIp);
-void    mgmtGetMnodeList(void *mpeers);
+void    mgmtGetMnodeList(void *mnodes);
 
 #ifdef __cplusplus
 }
diff --git a/src/mnode/inc/mgmtSdb.h b/src/mnode/inc/mgmtSdb.h
index 8ecb5ef1524060f3081e667ea132f934e37ba250..c09d215adb8d29ef24ccd3046c8acd47e977721d 100644
--- a/src/mnode/inc/mgmtSdb.h
+++ b/src/mnode/inc/mgmtSdb.h
@@ -36,54 +36,47 @@ typedef enum {
   SDB_KEY_STRING, 
   SDB_KEY_INT,
   SDB_KEY_AUTO
-} ESdbKeyType;
+} ESdbKey;
 
 typedef enum {
   SDB_OPER_GLOBAL,
   SDB_OPER_LOCAL
-} ESdbOperType;
+} ESdbOper;
 
 typedef struct {
-  ESdbOperType type;
-  void *  table;
-  void *  pObj;
-  int32_t rowSize;
-  void *  rowData;
-} SSdbOperDesc;
+  ESdbOper type;
+  void *   table;
+  void *   pObj;
+  int32_t  rowSize;
+  void *   rowData;
+} SSdbOper;
 
 typedef struct {
   char   *tableName;
   int32_t hashSessions;
   int32_t maxRowSize;
   int32_t refCountPos;
-  ESdbTable   tableId;
-  ESdbKeyType keyType;
-  int32_t (*insertFp)(SSdbOperDesc *pOper);
-  int32_t (*deleteFp)(SSdbOperDesc *pOper);
-  int32_t (*updateFp)(SSdbOperDesc *pOper);
-  int32_t (*encodeFp)(SSdbOperDesc *pOper);
-  int32_t (*decodeFp)(SSdbOperDesc *pDesc);  
-  int32_t (*destroyFp)(SSdbOperDesc *pDesc);
+  ESdbTable tableId;
+  ESdbKey   keyType;
+  int32_t (*insertFp)(SSdbOper *pOper);
+  int32_t (*deleteFp)(SSdbOper *pOper);
+  int32_t (*updateFp)(SSdbOper *pOper);
+  int32_t (*encodeFp)(SSdbOper *pOper);
+  int32_t (*decodeFp)(SSdbOper *pDesc);  
+  int32_t (*destroyFp)(SSdbOper *pDesc);
   int32_t (*restoredFp)();
 } SSdbTableDesc;
 
-typedef struct {
-  int64_t version;
-  void *  wal;
-  pthread_mutex_t mutex;
-} SSdbObject;
-
 int32_t sdbInit();
 void    sdbCleanUp();
-SSdbObject *sdbGetObj();
-
 void *  sdbOpenTable(SSdbTableDesc *desc);
 void    sdbCloseTable(void *handle);
-int     sdbProcessWrite(void *param, void *data, int type);
+bool    sdbIsMaster();
+void    sdbUpdateMnodeRoles();
 
-int32_t sdbInsertRow(SSdbOperDesc *pOper);
-int32_t sdbDeleteRow(SSdbOperDesc *pOper);
-int32_t sdbUpdateRow(SSdbOperDesc *pOper);
+int32_t sdbInsertRow(SSdbOper *pOper);
+int32_t sdbDeleteRow(SSdbOper *pOper);
+int32_t sdbUpdateRow(SSdbOper *pOper);
 
 void    *sdbGetRow(void *handle, void *key);
 void    *sdbFetchRow(void *handle, void *pNode, void **ppRow);
diff --git a/src/mnode/src/mgmtAcct.c b/src/mnode/src/mgmtAcct.c
index 3a52715274059f0b386810bcd8eb36a7f679b636..3e04399fe7f24384c9e459bd4adea27d7ee600e3 100644
--- a/src/mnode/src/mgmtAcct.c
+++ b/src/mnode/src/mgmtAcct.c
@@ -30,28 +30,28 @@ void *  tsAcctSdb = NULL;
 int32_t tsAcctUpdateSize;
 static void mgmtCreateRootAcct();
 
-static int32_t mgmtActionAcctDestroy(SSdbOperDesc *pOper) {
+static int32_t mgmtActionAcctDestroy(SSdbOper *pOper) {
   SAcctObj *pAcct = pOper->pObj;
   pthread_mutex_destroy(&pAcct->mutex);
   tfree(pOper->pObj);
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtAcctActionInsert(SSdbOperDesc *pOper) {
+static int32_t mgmtAcctActionInsert(SSdbOper *pOper) {
   SAcctObj *pAcct = pOper->pObj;
   memset(&pAcct->acctInfo, 0, sizeof(SAcctInfo));
   pthread_mutex_init(&pAcct->mutex, NULL);
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtActionAcctDelete(SSdbOperDesc *pOper) {
+static int32_t mgmtActionAcctDelete(SSdbOper *pOper) {
   SAcctObj *pAcct = pOper->pObj;
   mgmtDropAllUsers(pAcct);
   mgmtDropAllDbs(pAcct);
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtActionAcctUpdate(SSdbOperDesc *pOper) {
+static int32_t mgmtActionAcctUpdate(SSdbOper *pOper) {
   SAcctObj *pAcct = pOper->pObj;
   SAcctObj *pSaved = mgmtGetAcct(pAcct->user);
   if (pAcct != pSaved) {
@@ -61,14 +61,14 @@ static int32_t mgmtActionAcctUpdate(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtActionActionEncode(SSdbOperDesc *pOper) {
+static int32_t mgmtActionActionEncode(SSdbOper *pOper) {
   SAcctObj *pAcct = pOper->pObj;
   memcpy(pOper->rowData, pAcct, tsAcctUpdateSize);
   pOper->rowSize = tsAcctUpdateSize;
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtActionAcctDecode(SSdbOperDesc *pOper) {
+static int32_t mgmtActionAcctDecode(SSdbOper *pOper) {
   SAcctObj *pAcct = (SAcctObj *) calloc(1, sizeof(SAcctObj));
   if (pAcct == NULL) return TSDB_CODE_SERV_OUT_OF_MEMORY;
 
@@ -110,7 +110,7 @@ int32_t mgmtInitAccts() {
     return -1;
   }
 
-  mTrace("account table is created");
+  mTrace("table:accounts table is created");
   return acctInit();
 }
 
@@ -179,7 +179,7 @@ static void mgmtCreateRootAcct() {
   pAcct->acctId = sdbGetId(tsAcctSdb);
   pAcct->createdTime = taosGetTimestampMs();
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsAcctSdb,
     .pObj = pAcct,
diff --git a/src/mnode/src/mgmtReplica.c b/src/mnode/src/mgmtBalance.c
similarity index 86%
rename from src/mnode/src/mgmtReplica.c
rename to src/mnode/src/mgmtBalance.c
index 05a303a69bb82497672176b62897cf50d013a4bc..8ca651be2c6bca764752fdae7f20c89037e5f292 100644
--- a/src/mnode/src/mgmtReplica.c
+++ b/src/mnode/src/mgmtBalance.c
@@ -16,7 +16,7 @@
 #define _DEFAULT_SOURCE
 #include "os.h"
 #include "trpc.h"
-#include "treplica.h"
+#include "tbalance.h"
 #include "mgmtDef.h"
 #include "mgmtLog.h"
 #include "mgmtMnode.h"
@@ -25,13 +25,12 @@
 
 #ifndef _SYNC
 
-int32_t replicaInit() { return TSDB_CODE_SUCCESS; }
-void    replicaCleanUp() {}
-void    replicaNotify() {}
-void    replicaReset() {}
-int32_t replicaForwardReqToPeer(void *pHead) { return TSDB_CODE_SUCCESS; }
+int32_t balanceInit() { return TSDB_CODE_SUCCESS; }
+void    balanceCleanUp() {}
+void    balanceNotify() {}
+void    balanceReset() {}
 
-int32_t replicaAllocVnodes(SVgObj *pVgroup) {
+int32_t balanceAllocVnodes(SVgObj *pVgroup) {
   void *     pNode = NULL;
   SDnodeObj *pDnode = NULL;
   SDnodeObj *pSelDnode = NULL;
diff --git a/src/mnode/src/mgmtDServer.c b/src/mnode/src/mgmtDServer.c
index 4d8163deced31b5ab9260b43d3ffd07c025fe083..7d5b872f4a3949ed14197b217117829fdf7050a3 100644
--- a/src/mnode/src/mgmtDServer.c
+++ b/src/mnode/src/mgmtDServer.c
@@ -21,7 +21,7 @@
 #include "tsystem.h"
 #include "tutil.h"
 #include "tgrant.h"
-#include "treplica.h"
+#include "tbalance.h"
 #include "tglobalcfg.h"
 #include "dnode.h"
 #include "mgmtDef.h"
diff --git a/src/mnode/src/mgmtDb.c b/src/mnode/src/mgmtDb.c
index 9bf05ecfc4ac54e8e8ffaecd1fdfae77b1c3ba4c..90aaa03e9a9e495be597015a774fa06528733f10 100644
--- a/src/mnode/src/mgmtDb.c
+++ b/src/mnode/src/mgmtDb.c
@@ -46,12 +46,12 @@ static void    mgmtProcessCreateDbMsg(SQueuedMsg *pMsg);
 static void    mgmtProcessAlterDbMsg(SQueuedMsg *pMsg);
 static void    mgmtProcessDropDbMsg(SQueuedMsg *pMsg);
 
-static int32_t mgmtDbActionDestroy(SSdbOperDesc *pOper) {
+static int32_t mgmtDbActionDestroy(SSdbOper *pOper) {
   tfree(pOper->pObj);
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtDbActionInsert(SSdbOperDesc *pOper) {
+static int32_t mgmtDbActionInsert(SSdbOper *pOper) {
   SDbObj *pDb = pOper->pObj;
   SAcctObj *pAcct = mgmtGetAcct(pDb->cfg.acct);
 
@@ -72,7 +72,7 @@ static int32_t mgmtDbActionInsert(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtDbActionDelete(SSdbOperDesc *pOper) {
+static int32_t mgmtDbActionDelete(SSdbOper *pOper) {
   SDbObj *pDb = pOper->pObj;
   SAcctObj *pAcct = mgmtGetAcct(pDb->cfg.acct);
 
@@ -84,7 +84,7 @@ static int32_t mgmtDbActionDelete(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtDbActionUpdate(SSdbOperDesc *pOper) {
+static int32_t mgmtDbActionUpdate(SSdbOper *pOper) {
   SDbObj *pDb = pOper->pObj;
   SDbObj *pSaved = mgmtGetDb(pDb->name);
   if (pDb != pSaved) {
@@ -94,14 +94,14 @@ static int32_t mgmtDbActionUpdate(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtDbActionEncode(SSdbOperDesc *pOper) {
+static int32_t mgmtDbActionEncode(SSdbOper *pOper) {
   SDbObj *pDb = pOper->pObj;
   memcpy(pOper->rowData, pDb, tsDbUpdateSize);
   pOper->rowSize = tsDbUpdateSize;
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtDbActionDecode(SSdbOperDesc *pOper) {
+static int32_t mgmtDbActionDecode(SSdbOper *pOper) {
   SDbObj *pDb = (SDbObj *) calloc(1, sizeof(SDbObj));
   if (pDb == NULL) return TSDB_CODE_SERV_OUT_OF_MEMORY;
   
@@ -146,7 +146,7 @@ int32_t mgmtInitDbs() {
   mgmtAddShellShowMetaHandle(TSDB_MGMT_TABLE_DB, mgmtGetDbMeta);
   mgmtAddShellShowRetrieveHandle(TSDB_MGMT_TABLE_DB, mgmtRetrieveDbs);
   
-  mTrace("db data is initialized");
+  mTrace("table:dbs table is created");
   return 0;
 }
 
@@ -318,7 +318,7 @@ static int32_t mgmtCreateDb(SAcctObj *pAcct, SCMCreateDbMsg *pCreate) {
   pDb->createdTime = taosGetTimestampMs();
   pDb->cfg = *pCreate;
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsDbSdb,
     .pObj = pDb,
@@ -671,7 +671,7 @@ static int32_t mgmtSetDbDropping(SDbObj *pDb) {
   if (pDb->status) return TSDB_CODE_SUCCESS;
 
   pDb->status = true;
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsDbSdb,
     .pObj = pDb,
@@ -756,7 +756,7 @@ static int32_t mgmtAlterDb(SDbObj *pDb, SCMAlterDbMsg *pAlter) {
 
   if (memcmp(&newCfg, &pDb->cfg, sizeof(SDbCfg)) != 0) {
     pDb->cfg = newCfg;
-    SSdbOperDesc oper = {
+    SSdbOper oper = {
       .type = SDB_OPER_GLOBAL,
       .table = tsDbSdb,
       .pObj = pDb,
@@ -814,7 +814,7 @@ static void mgmtDropDb(SQueuedMsg *pMsg) {
   SDbObj *pDb = pMsg->pDb;
   mPrint("db:%s, drop db from sdb", pDb->name);
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsDbSdb,
     .pObj = pDb
diff --git a/src/mnode/src/mgmtDnode.c b/src/mnode/src/mgmtDnode.c
index 8b5969cfd019c18a7eda4d32d6da82e1eeb3c230..baec30942477f48fc6b35e6bf0655871d196a7a4 100644
--- a/src/mnode/src/mgmtDnode.c
+++ b/src/mnode/src/mgmtDnode.c
@@ -16,7 +16,7 @@
 #define _DEFAULT_SOURCE
 #include "os.h"
 #include "tgrant.h"
-#include "treplica.h"
+#include "tbalance.h"
 #include "tglobalcfg.h"
 #include "ttime.h"
 #include "tutil.h"
@@ -52,12 +52,12 @@ static int32_t mgmtRetrieveVnodes(SShowObj *pShow, char *data, int32_t rows, voi
 static int32_t mgmtGetDnodeMeta(STableMetaMsg *pMeta, SShowObj *pShow, void *pConn);
 static int32_t mgmtRetrieveDnodes(SShowObj *pShow, char *data, int32_t rows, void *pConn);
 
-static int32_t mgmtDnodeActionDestroy(SSdbOperDesc *pOper) {
+static int32_t mgmtDnodeActionDestroy(SSdbOper *pOper) {
   tfree(pOper->pObj);
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtDnodeActionInsert(SSdbOperDesc *pOper) {
+static int32_t mgmtDnodeActionInsert(SSdbOper *pOper) {
   SDnodeObj *pDnode = pOper->pObj;
   if (pDnode->status != TAOS_DN_STATUS_DROPPING) {
     pDnode->status = TAOS_DN_STATUS_OFFLINE;
@@ -72,7 +72,7 @@ static int32_t mgmtDnodeActionInsert(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtDnodeActionDelete(SSdbOperDesc *pOper) {
+static int32_t mgmtDnodeActionDelete(SSdbOper *pOper) {
   SDnodeObj *pDnode = pOper->pObj;
   void *     pNode = NULL;
   void *     pLastNode = NULL;
@@ -85,7 +85,7 @@ static int32_t mgmtDnodeActionDelete(SSdbOperDesc *pOper) {
     if (pVgroup == NULL) break;
 
     if (pVgroup->vnodeGid[0].dnodeId == pDnode->dnodeId) {
-      SSdbOperDesc oper = {
+      SSdbOper oper = {
         .type = SDB_OPER_LOCAL,
         .table = tsVgroupSdb,
         .pObj = pVgroup,
@@ -101,7 +101,7 @@ static int32_t mgmtDnodeActionDelete(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtDnodeActionUpdate(SSdbOperDesc *pOper) {
+static int32_t mgmtDnodeActionUpdate(SSdbOper *pOper) {
   SDnodeObj *pDnode = pOper->pObj;
   SDnodeObj *pSaved = mgmtGetDnode(pDnode->dnodeId);
   if (pDnode != pSaved) {
@@ -111,14 +111,14 @@ static int32_t mgmtDnodeActionUpdate(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtDnodeActionEncode(SSdbOperDesc *pOper) {
+static int32_t mgmtDnodeActionEncode(SSdbOper *pOper) {
   SDnodeObj *pDnode = pOper->pObj;
   memcpy(pOper->rowData, pDnode, tsDnodeUpdateSize);
   pOper->rowSize = tsDnodeUpdateSize;
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtDnodeActionDecode(SSdbOperDesc *pOper) {
+static int32_t mgmtDnodeActionDecode(SSdbOper *pOper) {
   SDnodeObj *pDnode = (SDnodeObj *) calloc(1, sizeof(SDnodeObj));
   if (pDnode == NULL) return TSDB_CODE_SERV_OUT_OF_MEMORY;
 
@@ -180,7 +180,7 @@ int32_t mgmtInitDnodes() {
   mgmtAddShellShowMetaHandle(TSDB_MGMT_TABLE_DNODE, mgmtGetDnodeMeta);
   mgmtAddShellShowRetrieveHandle(TSDB_MGMT_TABLE_DNODE, mgmtRetrieveDnodes);
  
-  mTrace("dnodes table is created");
+  mTrace("table:dnodes table is created");
   return 0;
 }
 
@@ -221,7 +221,7 @@ void mgmtReleaseDnode(SDnodeObj *pDnode) {
 }
 
 void mgmtUpdateDnode(SDnodeObj *pDnode) {
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsDnodeSdb,
     .pObj = pDnode,
@@ -340,7 +340,7 @@ void mgmtProcessDnodeStatusMsg(SRpcMsg *rpcMsg) {
   if (pDnode->status == TAOS_DN_STATUS_OFFLINE) {
     mTrace("dnode:%d, from offline to online", pDnode->dnodeId);
     pDnode->status = TAOS_DN_STATUS_READY;
-    replicaNotify();
+    balanceNotify();
   }
 
   mgmtReleaseDnode(pDnode);
@@ -393,7 +393,7 @@ static int32_t mgmtCreateDnode(uint32_t ip) {
   pDnode->totalVnodes = TSDB_INVALID_VNODE_NUM; 
   sprintf(pDnode->dnodeName, "n%d", sdbGetId(tsDnodeSdb) + 1);
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsDnodeSdb,
     .pObj = pDnode,
@@ -413,7 +413,7 @@ static int32_t mgmtCreateDnode(uint32_t ip) {
 }
 
 int32_t mgmtDropDnode(SDnodeObj *pDnode) {
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsDnodeSdb,
     .pObj = pDnode
diff --git a/src/mnode/src/mgmtMain.c b/src/mnode/src/mgmtMain.c
index 38f18b462ae69427088eea77bc7652ce9d144624..1d2f62a5935165ff9b6d4d0759a00882d36cbe5a 100644
--- a/src/mnode/src/mgmtMain.c
+++ b/src/mnode/src/mgmtMain.c
@@ -17,7 +17,7 @@
 #include "os.h"
 #include "taosdef.h"
 #include "tsched.h"
-#include "treplica.h"
+#include "tbalance.h"
 #include "tgrant.h"
 #include "ttimer.h"
 #include "dnode.h"
@@ -62,7 +62,7 @@ int32_t mgmtStartSystem() {
   }
 
   if (grantInit() < 0) {
-    mError("failed to init grants");
+    mError("failed to init grant");
     return -1;
   }
 
@@ -92,7 +92,7 @@ int32_t mgmtStartSystem() {
   }
 
   if (mgmtInitMnodes() < 0) {
-    mError("failed to init mpeers");
+    mError("failed to init mnodes");
     return -1;
   }
 
@@ -101,8 +101,8 @@ int32_t mgmtStartSystem() {
     return -1;
   }
 
-  if (replicaInit() < 0) {
-    mError("failed to init replica")
+  if (balanceInit() < 0) {
+    mError("failed to init balance")
   }
 
   if (mgmtInitDClient() < 0) {
@@ -144,7 +144,7 @@ void mgmtCleanUpSystem() {
   mPrint("starting to clean up mgmt");
   grantCleanUp();
   mgmtCleanupMnodes();
-  replicaCleanUp();
+  balanceCleanUp();
   mgmtCleanUpShell();
   mgmtCleanupDClient();
   mgmtCleanupDServer();
@@ -161,7 +161,7 @@ void mgmtCleanUpSystem() {
 }
 
 void mgmtStopSystem() {
-  if (mgmtIsMaster()) {
+  if (sdbIsMaster()) {
     mTrace("it is a master mgmt node, it could not be stopped");
     return;
   }
diff --git a/src/mnode/src/mgmtMnode.c b/src/mnode/src/mgmtMnode.c
index 8087ce5ad1513cdc89795fbe745bad28c96c96f6..bfb6480aec3eb8898b6bfd7b68255985999007b8 100644
--- a/src/mnode/src/mgmtMnode.c
+++ b/src/mnode/src/mgmtMnode.c
@@ -18,7 +18,7 @@
 #include "taoserror.h"
 #include "trpc.h"
 #include "tsync.h"
-#include "treplica.h"
+#include "tbalance.h"
 #include "tutil.h"
 #include "ttime.h"
 #include "tsocket.h"
@@ -30,18 +30,17 @@
 #include "mgmtShell.h"
 #include "mgmtUser.h"
 
-int32_t tsMnodeIsMaster = true;
 static void *  tsMnodeSdb = NULL;
 static int32_t tsMnodeUpdateSize = 0;
 static int32_t mgmtGetMnodeMeta(STableMetaMsg *pMeta, SShowObj *pShow, void *pConn);
 static int32_t mgmtRetrieveMnodes(SShowObj *pShow, char *data, int32_t rows, void *pConn);
 
-static int32_t mgmtMnodeActionDestroy(SSdbOperDesc *pOper) {
+static int32_t mgmtMnodeActionDestroy(SSdbOper *pOper) {
   tfree(pOper->pObj);
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtMnodeActionInsert(SSdbOperDesc *pOper) {
+static int32_t mgmtMnodeActionInsert(SSdbOper *pOper) {
   SMnodeObj *pMnode = pOper->pObj;
   SDnodeObj *pDnode = mgmtGetDnode(pMnode->mnodeId);
   if (pDnode == NULL) return TSDB_CODE_DNODE_NOT_EXIST;
@@ -53,7 +52,7 @@ static int32_t mgmtMnodeActionInsert(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtMnodeActionDelete(SSdbOperDesc *pOper) {
+static int32_t mgmtMnodeActionDelete(SSdbOper *pOper) {
   SMnodeObj *pMnode = pOper->pObj;
 
   SDnodeObj *pDnode = mgmtGetDnode(pMnode->mnodeId);
@@ -65,7 +64,7 @@ static int32_t mgmtMnodeActionDelete(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtMnodeActionUpdate(SSdbOperDesc *pOper) {
+static int32_t mgmtMnodeActionUpdate(SSdbOper *pOper) {
   SMnodeObj *pMnode = pOper->pObj;
   SMnodeObj *pSaved = mgmtGetMnode(pMnode->mnodeId);
   if (pMnode != pSaved) {
@@ -76,14 +75,14 @@ static int32_t mgmtMnodeActionUpdate(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtMnodeActionEncode(SSdbOperDesc *pOper) {
+static int32_t mgmtMnodeActionEncode(SSdbOper *pOper) {
   SMnodeObj *pMnode = pOper->pObj;
   memcpy(pOper->rowData, pMnode, tsMnodeUpdateSize);
   pOper->rowSize = tsMnodeUpdateSize;
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtMnodeActionDecode(SSdbOperDesc *pOper) {
+static int32_t mgmtMnodeActionDecode(SSdbOper *pOper) {
   SMnodeObj *pMnode = calloc(1, sizeof(SMnodeObj));
   if (pMnode == NULL) return TSDB_CODE_SERV_OUT_OF_MEMORY;
 
@@ -133,7 +132,7 @@ int32_t mgmtInitMnodes() {
   mgmtAddShellShowMetaHandle(TSDB_MGMT_TABLE_MNODE, mgmtGetMnodeMeta);
   mgmtAddShellShowRetrieveHandle(TSDB_MGMT_TABLE_MNODE, mgmtRetrieveMnodes);
 
-  mTrace("mnodes table is created");
+  mTrace("table:mnodes table is created");
   return TSDB_CODE_SUCCESS;
 }
 
@@ -157,7 +156,7 @@ void *mgmtGetNextMnode(void *pNode, SMnodeObj **pMnode) {
   return sdbFetchRow(tsMnodeSdb, pNode, (void **)pMnode); 
 }
 
-static char *mgmtGetMnodeRoleStr(int32_t role) {
+char *mgmtGetMnodeRoleStr(int32_t role) {
   switch (role) {
     case TAOS_SYNC_ROLE_OFFLINE:
       return "offline";
@@ -172,8 +171,6 @@ static char *mgmtGetMnodeRoleStr(int32_t role) {
   }
 }
 
-bool mgmtIsMaster() { return tsMnodeIsMaster; }
-
 void mgmtGetMnodeIpList(SRpcIpSet *ipSet, bool usePublicIp) {
   void *pNode = NULL;
   while (1) {
@@ -213,10 +210,8 @@ void mgmtGetMnodeList(void *param) {
     mnodes->nodeInfos[index].nodeIp = htonl(pMnode->pDnode->privateIp);
     mnodes->nodeInfos[index].nodePort = htons(pMnode->pDnode->mnodeDnodePort);
     strcpy(mnodes->nodeInfos[index].nodeName, pMnode->pDnode->dnodeName);
-    mPrint("node:%d role:%s", pMnode->mnodeId, mgmtGetMnodeRoleStr(pMnode->role));
     if (pMnode->role == TAOS_SYNC_ROLE_MASTER) {
       mnodes->inUse = index;
-      mPrint("node:%d inUse:%d", pMnode->mnodeId, mnodes->inUse);
     }
 
     index++;
@@ -231,7 +226,7 @@ int32_t mgmtAddMnode(int32_t dnodeId) {
   pMnode->mnodeId = dnodeId;
   pMnode->createdTime = taosGetTimestampMs();
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsMnodeSdb,
     .pObj = pMnode,
@@ -252,7 +247,7 @@ int32_t mgmtDropMnode(int32_t dnodeId) {
     return TSDB_CODE_DNODE_NOT_EXIST;
   }
   
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsMnodeSdb,
     .pObj = pMnode
@@ -268,6 +263,7 @@ int32_t mgmtDropMnode(int32_t dnodeId) {
 }
 
 static int32_t mgmtGetMnodeMeta(STableMetaMsg *pMeta, SShowObj *pShow, void *pConn) {
+  sdbUpdateMnodeRoles();
   SUserObj *pUser = mgmtGetUserFromConn(pConn, NULL);
   if (pUser == NULL) return 0;
 
diff --git a/src/mnode/src/mgmtSdb.c b/src/mnode/src/mgmtSdb.c
index 4bc18d6a0de369bd417f4e8f9e4abdf7cc439e17..3344b2b759e5e74d8eb099434aabb8639bf2ee6f 100644
--- a/src/mnode/src/mgmtSdb.c
+++ b/src/mnode/src/mgmtSdb.c
@@ -18,65 +18,93 @@
 #include "taoserror.h"
 #include "tlog.h"
 #include "trpc.h"
-#include "treplica.h"
+#include "tutil.h"
+#include "tbalance.h"
 #include "tqueue.h"
 #include "twal.h"
+#include "tsync.h"
 #include "hashint.h"
 #include "hashstr.h"
+#include "dnode.h"
+#include "mgmtDef.h"
 #include "mgmtLog.h"
 #include "mgmtMnode.h"
 #include "mgmtSdb.h"
 
+typedef enum {
+  SDB_ACTION_INSERT,
+  SDB_ACTION_DELETE,
+  SDB_ACTION_UPDATE
+} ESdbAction;
+
+typedef enum {
+  SDB_STATUS_OFFLINE,
+  SDB_STATUS_SERVING,
+  SDB_ACTION_CLOSING
+} ESdbStatus;
+
 typedef struct _SSdbTable {
-  char        tableName[TSDB_DB_NAME_LEN + 1];
-  ESdbTable   tableId;
-  ESdbKeyType keyType;
-  int32_t     hashSessions;
-  int32_t     maxRowSize;
-  int32_t     refCountPos;
-  int32_t     autoIndex;
-  int64_t     numOfRows;
-  void *      iHandle;
-  int32_t   (*insertFp)(SSdbOperDesc *pDesc);
-  int32_t   (*deleteFp)(SSdbOperDesc *pOper);
-  int32_t   (*updateFp)(SSdbOperDesc *pOper);
-  int32_t   (*decodeFp)(SSdbOperDesc *pOper);
-  int32_t   (*encodeFp)(SSdbOperDesc *pOper);
-  int32_t   (*destroyFp)(SSdbOperDesc *pOper);
-  int32_t   (*restoredFp)();
+  char      tableName[TSDB_DB_NAME_LEN + 1];
+  ESdbTable tableId;
+  ESdbKey   keyType;
+  int32_t   hashSessions;
+  int32_t   maxRowSize;
+  int32_t   refCountPos;
+  int32_t   autoIndex;
+  int64_t   numOfRows;
+  void *    iHandle;
+  int32_t (*insertFp)(SSdbOper *pDesc);
+  int32_t (*deleteFp)(SSdbOper *pOper);
+  int32_t (*updateFp)(SSdbOper *pOper);
+  int32_t (*decodeFp)(SSdbOper *pOper);
+  int32_t (*encodeFp)(SSdbOper *pOper);
+  int32_t (*destroyFp)(SSdbOper *pOper);
+  int32_t (*restoredFp)();
   pthread_mutex_t mutex;
 } SSdbTable;
 
+typedef struct {
+  ESyncRole  role;
+  ESdbStatus status;
+  int64_t    version;
+  void *     sync;
+  void *     wal;
+  SSyncCfg   cfg;
+  sem_t      sem;
+  int32_t    code;
+  int32_t    numOfTables;
+  SSdbTable *tableList[SDB_TABLE_MAX];
+  pthread_mutex_t mutex;
+} SSdbObject;
+
 typedef struct {
   int32_t rowSize;
   void *  row;
-} SRowMeta;
-
-typedef enum {
-  SDB_ACTION_INSERT,
-  SDB_ACTION_DELETE,
-  SDB_ACTION_UPDATE
-} ESdbActionType;
-
-static SSdbTable *tsSdbTableList[SDB_TABLE_MAX] = {0};
-static int32_t    tsSdbNumOfTables = 0;
-static SSdbObject * tsSdbObj;
+} SSdbRow;
 
+static SSdbObject tsSdbObj = {0};
 static void *(*sdbInitIndexFp[])(int32_t maxRows, int32_t dataSize) = {sdbOpenStrHash, sdbOpenIntHash, sdbOpenIntHash};
 static void *(*sdbAddIndexFp[])(void *handle, void *key, void *data) = {sdbAddStrHash, sdbAddIntHash, sdbAddIntHash};
 static void  (*sdbDeleteIndexFp[])(void *handle, void *key) = {sdbDeleteStrHash, sdbDeleteIntHash, sdbDeleteIntHash};
 static void *(*sdbGetIndexFp[])(void *handle, void *key) = {sdbGetStrHashData, sdbGetIntHashData, sdbGetIntHashData};
 static void  (*sdbCleanUpIndexFp[])(void *handle) = {sdbCloseStrHash, sdbCloseIntHash, sdbCloseIntHash};
 static void *(*sdbFetchRowFp[])(void *handle, void *ptr, void **ppRow) = {sdbFetchStrHashData, sdbFetchIntHashData, sdbFetchIntHashData};
+static int sdbWrite(void *param, void *data, int type);
 
-int32_t  sdbGetId(void *handle) { return ((SSdbTable *)handle)->autoIndex; }
-int64_t  sdbGetNumOfRows(void *handle) { return ((SSdbTable *)handle)->numOfRows; }
+int32_t sdbGetId(void *handle) {
+  return ((SSdbTable *)handle)->autoIndex;
+}
+
+int64_t sdbGetNumOfRows(void *handle) {
+  return ((SSdbTable *)handle)->numOfRows;
+}
 
 uint64_t sdbGetVersion() {
-  if (tsSdbObj)
-    return tsSdbObj->version;
-  else
-    return 0;
+  return tsSdbObj.version;
+}
+
+bool sdbIsMaster() { 
+  return tsSdbObj.role == TAOS_SYNC_ROLE_MASTER; 
 }
 
 static char *sdbGetActionStr(int32_t action) {
@@ -106,26 +134,26 @@ static char *sdbGetkeyStr(SSdbTable *pTable, void *row) {
 }
 
 static void *sdbGetTableFromId(int32_t tableId) {
-  return tsSdbTableList[tableId];
+  return tsSdbObj.tableList[tableId];
 }
 
-int32_t sdbInit() {
-  tsSdbObj = calloc(1, sizeof(SSdbObject));
-  pthread_mutex_init(&tsSdbObj->mutex, NULL);
-
+static int32_t sdbInitWal() {
   SWalCfg walCfg = {.commitLog = 2, .wals = 2, .keep = 1};
-  tsSdbObj->wal = walOpen(tsMnodeDir, &walCfg);
-  if (tsSdbObj->wal == NULL) {
-    sdbError("failed to open sdb in %s", tsMnodeDir);
+  tsSdbObj.wal = walOpen(tsMnodeDir, &walCfg);
+  if (tsSdbObj.wal == NULL) {
+    sdbError("failed to open sdb wal in %s", tsMnodeDir);
     return -1;
   }
 
-  sdbTrace("open sdb file for read");
-  walRestore(tsSdbObj->wal, tsSdbObj, sdbProcessWrite);
+  sdbTrace("open sdb wal for restore");
+  walRestore(tsSdbObj.wal, &tsSdbObj, sdbWrite);
+  return 0;
+}
 
+static void sdbRestoreTables() {
   int32_t totalRows = 0;
   int32_t numOfTables = 0;
-  for (int32_t tableId = SDB_TABLE_DNODE; tableId < SDB_TABLE_MAX; ++tableId) {
+  for (int32_t tableId = 0; tableId < SDB_TABLE_MAX; ++tableId) {
     SSdbTable *pTable = sdbGetTableFromId(tableId);
     if (pTable == NULL) continue;
     if (pTable->restoredFp) {
@@ -134,23 +162,170 @@ int32_t sdbInit() {
 
     totalRows += pTable->numOfRows;
     numOfTables++;
-    sdbTrace("table:%s, is initialized, numOfRows:%d", pTable->tableName, pTable->numOfRows);
+    sdbTrace("table:%s, is restored, numOfRows:%d", pTable->tableName, pTable->numOfRows);
+  }
+
+  sdbTrace("sdb is restored, version:%d totalRows:%d numOfTables:%d", tsSdbObj.version, totalRows, numOfTables);
+}
+
+void sdbUpdateMnodeRoles() {
+  if (tsSdbObj.sync == NULL) return;
+
+  SNodesRole roles = {0};
+  syncGetNodesRole(tsSdbObj.sync, &roles);
+
+  mPrint("update mnodes:%d sync roles", tsSdbObj.cfg.replica);
+  for (int32_t i = 0; i < tsSdbObj.cfg.replica; ++i) {
+    SMnodeObj *pMnode = mgmtGetMnode(roles.nodeId[i]);
+    if (pMnode != NULL) {
+      pMnode->role = roles.role[i];
+      mPrint("mnode:%d, role:%s", pMnode->mnodeId, mgmtGetMnodeRoleStr(pMnode->role));
+      mgmtReleaseMnode(pMnode);
+    }
+  }
+}
+
+static uint32_t sdbGetFileInfo(void *ahandle, char *name, uint32_t *index, int32_t *size) {
+  sdbUpdateMnodeRoles();
+  return 0;
+}
+
+static int sdbGetWalInfo(void *ahandle, char *name, uint32_t *index) {
+  strcpy(name, "wal0");
+  return 0;
+}
+
+static void sdbNotifyRole(void *ahandle, int8_t role) {
+  mPrint("mnode role changed from %s to %s", mgmtGetMnodeRoleStr(tsSdbObj.role), mgmtGetMnodeRoleStr(role));
+
+  if (role == TAOS_SYNC_ROLE_MASTER && tsSdbObj.role != TAOS_SYNC_ROLE_MASTER) {
+    balanceReset();
+  }
+  tsSdbObj.role = role;
+
+  sdbUpdateMnodeRoles();
+}
+
+static void sdbConfirmForward(void *ahandle, void *param, int32_t code) {
+  tsSdbObj.code = code;
+  sem_post(&tsSdbObj.sem);
+  mPrint("sdb forward request confirmed, result:%s", tstrerror(code));
+}
+
+static int32_t sdbForwardToPeer(void *pHead) {
+  if (tsSdbObj.sync == NULL) return TSDB_CODE_SUCCESS;
+
+  int32_t code = syncForwardToPeer(tsSdbObj.sync, pHead, NULL);
+  if (code > 0) {
+    sem_wait(&tsSdbObj.sem);
+    return tsSdbObj.code;
+  } 
+  return code;
+}
+
+void sdbUpdateSync() {
+  SSyncCfg syncCfg = {0};
+  int32_t index = 0;
+
+  SDMNodeInfos *mnodes = dnodeGetMnodeList();
+  for (int32_t i = 0; i < mnodes->nodeNum; ++i) {
+    SDMNodeInfo *node = &mnodes->nodeInfos[i];
+    syncCfg.nodeInfo[i].nodeId = node->nodeId;
+    syncCfg.nodeInfo[i].nodeIp = node->nodeIp;
+    strcpy(syncCfg.nodeInfo[i].name, node->nodeName);
+    index++;
+  }
+
+  if (index == 0) {
+    void *pNode = NULL;
+    while (1) {
+      SMnodeObj *pMnode = NULL;
+      pNode = mgmtGetNextMnode(pNode, &pMnode);
+      if (pMnode == NULL) break;
+
+      syncCfg.nodeInfo[index].nodeId = pMnode->mnodeId;
+      syncCfg.nodeInfo[index].nodeIp = pMnode->pDnode->privateIp;
+      strcpy(syncCfg.nodeInfo[index].name, pMnode->pDnode->dnodeName);
+      index++;
+
+      mgmtReleaseMnode(pMnode);
+    }
+  }
+
+  syncCfg.replica = index;
+  syncCfg.arbitratorIp = syncCfg.nodeInfo[0].nodeIp;
+  if (syncCfg.replica == 1) {
+    syncCfg.quorum = 1;
+  } else {
+    syncCfg.quorum = 2;
+  }
+
+  bool hasThisDnode = false;
+  for (int32_t i = 0; i < syncCfg.replica; ++i) {
+    if (syncCfg.nodeInfo[i].nodeId == dnodeGetDnodeId()) {
+      hasThisDnode = true;
+      break;
+    }
+  }
+
+  if (!hasThisDnode) return;
+  if (memcmp(&syncCfg, &tsSdbObj.cfg, sizeof(SSyncCfg)) == 0) return;
+
+  mPrint("work as mnode, replica:%d arbitratorIp:%s", syncCfg.replica, taosIpStr(syncCfg.arbitratorIp));
+  for (int32_t i = 0; i < syncCfg.replica; ++i) {
+    mPrint("mnode:%d, ip:%s name:%s", syncCfg.nodeInfo[i].nodeId, taosIpStr(syncCfg.nodeInfo[i].nodeIp),
+           syncCfg.nodeInfo[i].name);
   }
 
-  sdbTrace("sdb is initialized, version:%d totalRows:%d numOfTables:%d", tsSdbObj->version, totalRows, numOfTables);
+  SSyncInfo syncInfo;
+  syncInfo.vgId = 1;
+  syncInfo.version = sdbGetVersion();
+  syncInfo.syncCfg = syncCfg;
+  sprintf(syncInfo.path, "%s/", tsMnodeDir);
+  syncInfo.ahandle = NULL;
+  syncInfo.getWalInfo = sdbGetWalInfo;
+  syncInfo.getFileInfo = sdbGetFileInfo;
+  syncInfo.writeToCache = sdbWrite;
+  syncInfo.confirmForward = sdbConfirmForward; 
+  syncInfo.notifyRole = sdbNotifyRole;
+  tsSdbObj.cfg = syncCfg;
+
+  if (tsSdbObj.sync) {
+    syncReconfig(tsSdbObj.sync, &syncCfg);
+  } else {
+    tsSdbObj.sync = syncStart(&syncInfo);
+  }
+}
+
+int32_t sdbInit() {
+  pthread_mutex_init(&tsSdbObj.mutex, NULL);
+  sem_init(&tsSdbObj.sem, 0, 0);
+
+  if (sdbInitWal() != 0) {
+    return -1;
+  }
   
-  replicaNotify();
+  sdbRestoreTables();
 
+  if (mgmtGetMnodesNum() == 1) {
+    tsSdbObj.role = TAOS_SYNC_ROLE_MASTER;
+  }
+
+  sdbUpdateSync();
+
+  tsSdbObj.status = SDB_STATUS_SERVING;
   return TSDB_CODE_SUCCESS;
 }
 
 void sdbCleanUp() {
-  if (tsSdbObj) {
-    pthread_mutex_destroy(&tsSdbObj->mutex);
-    walClose(tsSdbObj->wal);
-    free(tsSdbObj);
-    tsSdbObj = NULL;
-  }
+  if (tsSdbObj.status != SDB_STATUS_SERVING) return;
+
+  syncStop(tsSdbObj.sync);
+  free(tsSdbObj.sync);
+  walClose(tsSdbObj.wal);
+  sem_destroy(&tsSdbObj.sem);
+  pthread_mutex_destroy(&tsSdbObj.mutex);
+  memset(&tsSdbObj, 0, sizeof(tsSdbObj));
 }
 
 void sdbIncRef(void *handle, void *pRow) {
@@ -178,15 +353,15 @@ void sdbDecRef(void *handle, void *pRow) {
     if (refCount <= 0 && *updateEnd) {
       sdbTrace("table:%s, record:%s:%s:%d is destroyed", pTable->tableName, pTable->tableName,
                sdbGetkeyStr(pTable, pRow), *pRefCount);
-      SSdbOperDesc oper = {.pObj = pRow};
+      SSdbOper oper = {.pObj = pRow};
       (*pTable->destroyFp)(&oper);
     }
   }
 }
 
-static SRowMeta *sdbGetRowMeta(void *handle, void *key) {
+static SSdbRow *sdbGetRowMeta(void *handle, void *key) {
   SSdbTable *pTable = (SSdbTable *)handle;
-  SRowMeta * pMeta;
+  SSdbRow * pMeta;
 
   if (handle == NULL) return NULL;
 
@@ -197,7 +372,7 @@ static SRowMeta *sdbGetRowMeta(void *handle, void *key) {
 
 void *sdbGetRow(void *handle, void *key) {
   SSdbTable *pTable = (SSdbTable *)handle;
-  SRowMeta * pMeta;
+  SSdbRow * pMeta;
 
   if (handle == NULL) return NULL;
 
@@ -213,8 +388,8 @@ void *sdbGetRow(void *handle, void *key) {
   return pMeta->row;
 }
 
-static int32_t sdbInsertLocal(SSdbTable *pTable, SSdbOperDesc *pOper) {
-  SRowMeta rowMeta;
+static int32_t sdbInsertHash(SSdbTable *pTable, SSdbOper *pOper) {
+  SSdbRow rowMeta;
   rowMeta.rowSize = pOper->rowSize;
   rowMeta.row = pOper->pObj;
 
@@ -229,20 +404,20 @@ static int32_t sdbInsertLocal(SSdbTable *pTable, SSdbOperDesc *pOper) {
 
   pthread_mutex_unlock(&pTable->mutex);
 
-  sdbTrace("table:%s, insert record:%s, numOfRows:%d", pTable->tableName, sdbGetkeyStr(pTable, pOper->pObj),
+  sdbTrace("table:%s, insert record:%s to hash, numOfRows:%d", pTable->tableName, sdbGetkeyStr(pTable, pOper->pObj),
            pTable->numOfRows);
 
   (*pTable->insertFp)(pOper);
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t sdbDeleteLocal(SSdbTable *pTable, SSdbOperDesc *pOper) {
+static int32_t sdbDeleteHash(SSdbTable *pTable, SSdbOper *pOper) {
   pthread_mutex_lock(&pTable->mutex);
   (*sdbDeleteIndexFp[pTable->keyType])(pTable->iHandle, pOper->pObj);
   pTable->numOfRows--;
   pthread_mutex_unlock(&pTable->mutex);
 
-  sdbTrace("table:%s, delete record:%s, numOfRows:%d", pTable->tableName, sdbGetkeyStr(pTable, pOper->pObj),
+  sdbTrace("table:%s, delete record:%s from hash, numOfRows:%d", pTable->tableName, sdbGetkeyStr(pTable, pOper->pObj),
            pTable->numOfRows);
 
   (*pTable->deleteFp)(pOper);
@@ -253,127 +428,76 @@ static int32_t sdbDeleteLocal(SSdbTable *pTable, SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t sdbUpdateLocal(SSdbTable *pTable, SSdbOperDesc *pOper) {
-  sdbTrace("table:%s, update record:%s, numOfRows:%d", pTable->tableName, sdbGetkeyStr(pTable, pOper->pObj),
+static int32_t sdbUpdateHash(SSdbTable *pTable, SSdbOper *pOper) {
+  sdbTrace("table:%s, update record:%s in hash, numOfRows:%d", pTable->tableName, sdbGetkeyStr(pTable, pOper->pObj),
            pTable->numOfRows);
 
   (*pTable->updateFp)(pOper);
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t sdbProcessWriteFromApp(SSdbTable *pTable, SWalHead *pHead, int32_t action) {
-  int32_t code = 0;
+static int sdbWrite(void *param, void *data, int type) {
+  SWalHead *pHead = data;
+  int32_t   tableId = pHead->msgType / 10;
+  int32_t   action = pHead->msgType % 10;
 
-  pthread_mutex_lock(&tsSdbObj->mutex);
-  tsSdbObj->version++;
-  pHead->version = tsSdbObj->version;
+  SSdbTable *pTable = sdbGetTableFromId(tableId);
+  assert(pTable != NULL);
 
-  code = replicaForwardReqToPeer(pHead);
-  if (code != TSDB_CODE_SUCCESS) {
-    pthread_mutex_unlock(&tsSdbObj->mutex);
-    sdbError("table:%s, failed to forward %s record:%s from file, version:%" PRId64 ", reason:%s", pTable->tableName,
-             sdbGetActionStr(action), sdbGetkeyStr(pTable, pHead->cont), pHead->version, tstrerror(code));
-    return code;
+  pthread_mutex_lock(&tsSdbObj.mutex);
+  if (pHead->version == 0) {
+     // assign version
+    tsSdbObj.version++;
+    pHead->version = tsSdbObj.version;
+  } else {
+    // for data from WAL or forward, version may be smaller
+    if (pHead->version <= tsSdbObj.version) {
+      pthread_mutex_unlock(&tsSdbObj.mutex);
+      return TSDB_CODE_SUCCESS;
+    } else if (pHead->version != tsSdbObj.version + 1) {
+      pthread_mutex_unlock(&tsSdbObj.mutex);
+      sdbError("table:%s, failed to restore %s record:%s from wal, version:%" PRId64 " too large, sdb version:%" PRId64,
+               pTable->tableName, sdbGetActionStr(action), sdbGetkeyStr(pTable, pHead->cont), pHead->version,
+               tsSdbObj.version);
+      return TSDB_CODE_OTHERS;
+    } else {
+      tsSdbObj.version = pHead->version;
+    }
   }
 
-  code = walWrite(tsSdbObj->wal, pHead);
-  pthread_mutex_unlock(&tsSdbObj->mutex);
-
+  int32_t code = walWrite(tsSdbObj.wal, pHead);
   if (code < 0) {
-    sdbError("table:%s, failed to %s record:%s to file, version:%" PRId64 ", reason:%s", pTable->tableName,
-             sdbGetActionStr(action), sdbGetkeyStr(pTable, pHead->cont), pHead->version, tstrerror(code));
-  } else {
-    sdbTrace("table:%s, success to %s record:%s to file, version:%" PRId64, pTable->tableName, sdbGetActionStr(action),
-             sdbGetkeyStr(pTable, pHead->cont), pHead->version);
+    pthread_mutex_unlock(&tsSdbObj.mutex);
+    return code;
   }
+  walFsync(tsSdbObj.wal);
 
-  walFsync(tsSdbObj->wal);
-  taosFreeQitem(pHead);
-  return code;
-}
-
-static int32_t sdbProcessWriteFromWal(SSdbTable *pTable, SWalHead *pHead, int32_t action) {
-  pthread_mutex_lock(&tsSdbObj->mutex);
-  if (pHead->version <= tsSdbObj->version) {
-    pthread_mutex_unlock(&tsSdbObj->mutex);
-    return TSDB_CODE_SUCCESS;
-  } else if (pHead->version != tsSdbObj->version + 1) {
-    pthread_mutex_unlock(&tsSdbObj->mutex);
-    sdbError("table:%s, failed to restore %s record:%s from file, version:%" PRId64 " too large, sdb version:%" PRId64,
-             pTable->tableName, sdbGetActionStr(action), sdbGetkeyStr(pTable, pHead->cont), pHead->version,
-             tsSdbObj->version);
-    return TSDB_CODE_OTHERS;
-  }
+  sdbForwardToPeer(pHead);
+  pthread_mutex_unlock(&tsSdbObj.mutex);
 
-  tsSdbObj->version = pHead->version;
-  sdbTrace("table:%s, success to restore %s record:%s from file, version:%" PRId64, pTable->tableName,
-           sdbGetActionStr(action), sdbGetkeyStr(pTable, pHead->cont), pHead->version);
+  // from app, oper is created
+  if (param == NULL) return code;
 
-  int32_t code = -1;
+  // from wal, should create oper
   if (action == SDB_ACTION_INSERT) {
-    SSdbOperDesc oper = {
-        .rowSize = pHead->len,
-        .rowData = pHead->cont,
-        .table = pTable,
-    };
+    SSdbOper oper = {.rowSize = pHead->len, .rowData = pHead->cont, .table = pTable};
     code = (*pTable->decodeFp)(&oper);
-    if (code < 0) {
-      sdbTrace("table:%s, failed to decode %s record:%s from file, version:%" PRId64, pTable->tableName,
-               sdbGetActionStr(action), sdbGetkeyStr(pTable, pHead->cont), pHead->version);
-      pthread_mutex_unlock(&tsSdbObj->mutex);
-      return code;
-    }
-
-    code = sdbInsertLocal(pTable, &oper);
+    return sdbInsertHash(pTable, &oper);
   } else if (action == SDB_ACTION_DELETE) {
-    SRowMeta *rowMeta = sdbGetRowMeta(pTable, pHead->cont);
+    SSdbRow *rowMeta = sdbGetRowMeta(pTable, pHead->cont);
     assert(rowMeta != NULL && rowMeta->row != NULL);
-
-    SSdbOperDesc oper = {
-        .table = pTable,
-        .pObj = rowMeta->row,
-    };
-
-    code = sdbDeleteLocal(pTable, &oper);
+    SSdbOper oper = {.table = pTable, .pObj = rowMeta->row};
+    return sdbDeleteHash(pTable, &oper);
   } else if (action == SDB_ACTION_UPDATE) {
-    SRowMeta *rowMeta = sdbGetRowMeta(pTable, pHead->cont);
+    SSdbRow *rowMeta = sdbGetRowMeta(pTable, pHead->cont);
     assert(rowMeta != NULL && rowMeta->row != NULL);
-
-    SSdbOperDesc oper = {
-        .rowSize = pHead->len,
-        .rowData = pHead->cont,
-        .table = pTable,
-    };
+    SSdbOper oper = {.rowSize = pHead->len, .rowData = pHead->cont, .table = pTable};
     code = (*pTable->decodeFp)(&oper);
-    if (code < 0) {
-      sdbTrace("table:%s, failed to decode %s record:%s from file, version:%" PRId64, pTable->tableName,
-               sdbGetActionStr(action), sdbGetkeyStr(pTable, pHead->cont), pHead->version);
-      pthread_mutex_unlock(&tsSdbObj->mutex);
-      return code;
-    }
-    code = sdbUpdateLocal(pTable, &oper);
-  }
-
-  pthread_mutex_unlock(&tsSdbObj->mutex);
-  return code;
+    return sdbUpdateHash(pTable, &oper);
+  } else { return TSDB_CODE_INVALID_MSG_TYPE; }
 }
 
-int sdbProcessWrite(void *param, void *data, int type) {
-  SWalHead *pHead = data;
-  int32_t   tableId = pHead->msgType / 10;
-  int32_t   action = pHead->msgType % 10;
-
-  SSdbTable *pTable = sdbGetTableFromId(tableId);
-  assert(pTable != NULL);
-
-  if (pHead->version == 0) {
-    return sdbProcessWriteFromApp(pTable, pHead, action);
-  } else {
-    return sdbProcessWriteFromWal(pTable, pHead, action);
-  }
-}
-
-int32_t sdbInsertRow(SSdbOperDesc *pOper) {
+int32_t sdbInsertRow(SSdbOper *pOper) {
   SSdbTable *pTable = (SSdbTable *)pOper->table;
   if (pTable == NULL) return -1;
 
@@ -405,19 +529,19 @@ int32_t sdbInsertRow(SSdbOperDesc *pOper) {
     (*pTable->encodeFp)(pOper);
     pHead->len = pOper->rowSize;
 
-    int32_t code = sdbProcessWrite(tsSdbObj, pHead, pHead->msgType);
+    int32_t code = sdbWrite(NULL, pHead, pHead->msgType);
+    taosFreeQitem(pHead);
     if (code < 0) return code;
-  } 
-  
-  return sdbInsertLocal(pTable, pOper);
+  }
+
+  return sdbInsertHash(pTable, pOper);
 }
 
-// row here can be object or null-terminated string
-int32_t sdbDeleteRow(SSdbOperDesc *pOper) {
+int32_t sdbDeleteRow(SSdbOper *pOper) {
   SSdbTable *pTable = (SSdbTable *)pOper->table;
   if (pTable == NULL) return -1;
 
-  SRowMeta *pMeta = sdbGetRowMeta(pTable, pOper->pObj);
+  SSdbRow *pMeta = sdbGetRowMeta(pTable, pOper->pObj);
   if (pMeta == NULL) {
     sdbTrace("table:%s, record is not there, delete failed", pTable->tableName);
     return -1;
@@ -447,18 +571,19 @@ int32_t sdbDeleteRow(SSdbOperDesc *pOper) {
     pHead->msgType = pTable->tableId * 10 + SDB_ACTION_DELETE;
     memcpy(pHead->cont, pOper->pObj, rowSize);
 
-    int32_t code = sdbProcessWrite(tsSdbObj, pHead, pHead->msgType);
+    int32_t code = sdbWrite(NULL, pHead, pHead->msgType);
+    taosFreeQitem(pHead);
     if (code < 0) return code;
-  } 
-  
-  return sdbDeleteLocal(pTable, pOper);
+  }
+
+  return sdbDeleteHash(pTable, pOper);
 }
 
-int32_t sdbUpdateRow(SSdbOperDesc *pOper) {
+int32_t sdbUpdateRow(SSdbOper *pOper) {
   SSdbTable *pTable = (SSdbTable *)pOper->table;
   if (pTable == NULL) return -1;
 
-  SRowMeta *pMeta = sdbGetRowMeta(pTable, pOper->pObj);
+  SSdbRow *pMeta = sdbGetRowMeta(pTable, pOper->pObj);
   if (pMeta == NULL) {
     sdbTrace("table:%s, record is not there, delete failed", pTable->tableName);
     return -1;
@@ -477,16 +602,17 @@ int32_t sdbUpdateRow(SSdbOperDesc *pOper) {
     (*pTable->encodeFp)(pOper);
     pHead->len = pOper->rowSize;
 
-    int32_t code = sdbProcessWrite(tsSdbObj, pHead, pHead->msgType);
+    int32_t code = sdbWrite(NULL, pHead, pHead->msgType);
+    taosFreeQitem(pHead);
     if (code < 0) return code;
-  }
+  } 
   
-  return sdbUpdateLocal(pTable, pOper);
+  return sdbUpdateHash(pTable, pOper);
 }
 
 void *sdbFetchRow(void *handle, void *pNode, void **ppRow) {
   SSdbTable *pTable = (SSdbTable *)handle;
-  SRowMeta * pMeta;
+  SSdbRow * pMeta;
 
   *ppRow = NULL;
   if (pTable == NULL) return NULL;
@@ -520,13 +646,13 @@ void *sdbOpenTable(SSdbTableDesc *pDesc) {
   pTable->restoredFp   = pDesc->restoredFp;
   
   if (sdbInitIndexFp[pTable->keyType] != NULL) {
-    pTable->iHandle = (*sdbInitIndexFp[pTable->keyType])(pTable->maxRowSize, sizeof(SRowMeta));
+    pTable->iHandle = (*sdbInitIndexFp[pTable->keyType])(pTable->maxRowSize, sizeof(SSdbRow));
   }
 
   pthread_mutex_init(&pTable->mutex, NULL);
 
-  tsSdbNumOfTables++;
-  tsSdbTableList[pTable->tableId] = pTable;
+  tsSdbObj.numOfTables++;
+  tsSdbObj.tableList[pTable->tableId] = pTable;
   return pTable;
 }
 
@@ -534,16 +660,16 @@ void sdbCloseTable(void *handle) {
   SSdbTable *pTable = (SSdbTable *)handle;
   if (pTable == NULL) return;
   
-  tsSdbNumOfTables--;
-  tsSdbTableList[pTable->tableId] = NULL;
+  tsSdbObj.numOfTables--;
+  tsSdbObj.tableList[pTable->tableId] = NULL;
 
   void *pNode = NULL;
   while (1) {
-    SRowMeta *pMeta;
+    SSdbRow *pMeta;
     pNode = (*sdbFetchRowFp[pTable->keyType])(pTable->iHandle, pNode, (void **)&pMeta);
     if (pMeta == NULL) break;
 
-    SSdbOperDesc oper = {
+    SSdbOper oper = {
       .pObj = pMeta->row,
       .table = pTable,
     };
@@ -557,6 +683,7 @@ void sdbCloseTable(void *handle) {
 
   pthread_mutex_destroy(&pTable->mutex);
   
-  sdbTrace("table:%s, is closed, numOfTables:%d", pTable->tableName, tsSdbNumOfTables);
+  sdbTrace("table:%s, is closed, numOfTables:%d", pTable->tableName, tsSdbObj.numOfTables);
   free(pTable);
 }
+
diff --git a/src/mnode/src/mgmtShell.c b/src/mnode/src/mgmtShell.c
index 1dec4bd0153ee27299e946da933460eb7c5d802f..04f771081be058dfa15174dd522d3ac608c11c21 100644
--- a/src/mnode/src/mgmtShell.c
+++ b/src/mnode/src/mgmtShell.c
@@ -144,7 +144,7 @@ static void mgmtProcessMsgFromShell(SRpcMsg *rpcMsg) {
     return;
   }
 
-  if (!mgmtIsMaster()) {
+  if (!sdbIsMaster()) {
     // rpcSendRedirectRsp(rpcMsg->handle, mgmtGetMnodeIpListForRedirect());
     mgmtSendSimpleResp(rpcMsg->handle, TSDB_CODE_NO_MASTER);
     rpcFreeCont(rpcMsg->pCont);
diff --git a/src/mnode/src/mgmtTable.c b/src/mnode/src/mgmtTable.c
index 3fb4272b7f36f61ee31a2bb9029a03c3ac39e995..383b3474102c89f208f5a2860733c4f1e58e299e 100644
--- a/src/mnode/src/mgmtTable.c
+++ b/src/mnode/src/mgmtTable.c
@@ -83,12 +83,12 @@ static void mgmtDestroyChildTable(SChildTableObj *pTable) {
   tfree(pTable);
 }
 
-static int32_t mgmtChildTableActionDestroy(SSdbOperDesc *pOper) {
+static int32_t mgmtChildTableActionDestroy(SSdbOper *pOper) {
   mgmtDestroyChildTable(pOper->pObj);
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtChildTableActionInsert(SSdbOperDesc *pOper) {
+static int32_t mgmtChildTableActionInsert(SSdbOper *pOper) {
   SChildTableObj *pTable = pOper->pObj;
 
   SVgObj *pVgroup = mgmtGetVgroup(pTable->vgId);
@@ -128,7 +128,7 @@ static int32_t mgmtChildTableActionInsert(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtChildTableActionDelete(SSdbOperDesc *pOper) {
+static int32_t mgmtChildTableActionDelete(SSdbOper *pOper) {
   SChildTableObj *pTable = pOper->pObj;
   if (pTable->vgId == 0) {
     return TSDB_CODE_INVALID_VGROUP_ID;
@@ -169,7 +169,7 @@ static int32_t mgmtChildTableActionDelete(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtChildTableActionUpdate(SSdbOperDesc *pOper) {
+static int32_t mgmtChildTableActionUpdate(SSdbOper *pOper) {
   SChildTableObj *pNew = pOper->pObj;
   SChildTableObj *pTable = mgmtGetChildTable(pNew->info.tableId);
   if (pTable != pNew) {
@@ -186,7 +186,7 @@ static int32_t mgmtChildTableActionUpdate(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtChildTableActionEncode(SSdbOperDesc *pOper) {
+static int32_t mgmtChildTableActionEncode(SSdbOper *pOper) {
   const int32_t maxRowSize = sizeof(SChildTableObj) + sizeof(SSchema) * TSDB_MAX_COLUMNS;
   SChildTableObj *pTable = pOper->pObj;
   assert(pTable != NULL && pOper->rowData != NULL);
@@ -208,7 +208,7 @@ static int32_t mgmtChildTableActionEncode(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtChildTableActionDecode(SSdbOperDesc *pOper) {
+static int32_t mgmtChildTableActionDecode(SSdbOper *pOper) {
   assert(pOper->rowData != NULL);
   SChildTableObj *pTable = calloc(1, sizeof(SChildTableObj));
   if (pTable == NULL) {
@@ -252,7 +252,7 @@ static int32_t mgmtChildTableActionRestored() {
     SDbObj *pDb = mgmtGetDbByTableId(pTable->info.tableId);
     if (pDb == NULL) {
       mError("ctable:%s, failed to get db, discard it", pTable->info.tableId);
-      SSdbOperDesc desc = {0};
+      SSdbOper desc = {0};
       desc.type = SDB_OPER_LOCAL;
       desc.pObj = pTable;
       desc.table = tsChildTableSdb;
@@ -266,7 +266,7 @@ static int32_t mgmtChildTableActionRestored() {
     if (pVgroup == NULL) {
       mError("ctable:%s, failed to get vgroup:%d sid:%d, discard it", pTable->info.tableId, pTable->vgId, pTable->sid);
       pTable->vgId = 0;
-      SSdbOperDesc desc = {0};
+      SSdbOper desc = {0};
       desc.type = SDB_OPER_LOCAL;
       desc.pObj = pTable;
       desc.table = tsChildTableSdb;
@@ -280,7 +280,7 @@ static int32_t mgmtChildTableActionRestored() {
       mError("ctable:%s, db:%s not match with vgroup:%d db:%s sid:%d, discard it",
              pTable->info.tableId, pDb->name, pTable->vgId, pVgroup->dbName, pTable->sid);
       pTable->vgId = 0;
-      SSdbOperDesc desc = {0};
+      SSdbOper desc = {0};
       desc.type = SDB_OPER_LOCAL;
       desc.pObj = pTable;
       desc.table = tsChildTableSdb;
@@ -292,7 +292,7 @@ static int32_t mgmtChildTableActionRestored() {
     if (pVgroup->tableList == NULL) {
       mError("ctable:%s, vgroup:%d tableList is null", pTable->info.tableId, pTable->vgId);
       pTable->vgId = 0;
-      SSdbOperDesc desc = {0};
+      SSdbOper desc = {0};
       desc.type = SDB_OPER_LOCAL;
       desc.pObj = pTable;
       desc.table = tsChildTableSdb;
@@ -306,7 +306,7 @@ static int32_t mgmtChildTableActionRestored() {
       if (pSuperTable == NULL) {
         mError("ctable:%s, stable:%s not exist", pTable->info.tableId, pTable->superTableId);
         pTable->vgId = 0;
-        SSdbOperDesc desc = {0};
+        SSdbOper desc = {0};
         desc.type = SDB_OPER_LOCAL;
         desc.pObj = pTable;
         desc.table = tsChildTableSdb;
@@ -347,7 +347,7 @@ static int32_t mgmtInitChildTables() {
     return -1;
   }
 
-  mTrace("child table is initialized");
+  mTrace("table:ctables is created");
   return 0;
 }
 
@@ -392,12 +392,12 @@ static void mgmtDestroySuperTable(SSuperTableObj *pStable) {
   tfree(pStable);
 }
 
-static int32_t mgmtSuperTableActionDestroy(SSdbOperDesc *pOper) {
+static int32_t mgmtSuperTableActionDestroy(SSdbOper *pOper) {
   mgmtDestroySuperTable(pOper->pObj);
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtSuperTableActionInsert(SSdbOperDesc *pOper) {
+static int32_t mgmtSuperTableActionInsert(SSdbOper *pOper) {
   SSuperTableObj *pStable = pOper->pObj;
   SDbObj *pDb = mgmtGetDbByTableId(pStable->info.tableId);
   if (pDb != NULL) {
@@ -408,7 +408,7 @@ static int32_t mgmtSuperTableActionInsert(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtSuperTableActionDelete(SSdbOperDesc *pOper) {
+static int32_t mgmtSuperTableActionDelete(SSdbOper *pOper) {
   SSuperTableObj *pStable = pOper->pObj;
   SDbObj *pDb = mgmtGetDbByTableId(pStable->info.tableId);
   if (pDb != NULL) {
@@ -420,7 +420,7 @@ static int32_t mgmtSuperTableActionDelete(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtSuperTableActionUpdate(SSdbOperDesc *pOper) {
+static int32_t mgmtSuperTableActionUpdate(SSdbOper *pOper) {
   SSuperTableObj *pNew = pOper->pObj;
   SSuperTableObj *pTable = mgmtGetSuperTable(pNew->info.tableId);
   if (pTable != pNew) {
@@ -435,7 +435,7 @@ static int32_t mgmtSuperTableActionUpdate(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtSuperTableActionEncode(SSdbOperDesc *pOper) {
+static int32_t mgmtSuperTableActionEncode(SSdbOper *pOper) {
   const int32_t maxRowSize = sizeof(SChildTableObj) + sizeof(SSchema) * TSDB_MAX_COLUMNS;
 
   SSuperTableObj *pStable = pOper->pObj;
@@ -454,7 +454,7 @@ static int32_t mgmtSuperTableActionEncode(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtSuperTableActionDecode(SSdbOperDesc *pOper) {
+static int32_t mgmtSuperTableActionDecode(SSdbOper *pOper) {
   assert(pOper->rowData != NULL);
 
   SSuperTableObj *pStable = (SSuperTableObj *) calloc(1, sizeof(SSuperTableObj));
@@ -505,7 +505,7 @@ static int32_t mgmtInitSuperTables() {
     return -1;
   }
 
-  mTrace("stables is initialized");
+  mTrace("table:stables is created");
   return 0;
 }
 
@@ -731,7 +731,7 @@ static void mgmtProcessCreateSuperTableMsg(SQueuedMsg *pMsg) {
     tschema[col].bytes = htons(tschema[col].bytes);
   }
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsSuperTableSdb,
     .pObj = pStable,
@@ -755,7 +755,7 @@ static void mgmtProcessDropSuperTableMsg(SQueuedMsg *pMsg) {
     mError("stable:%s, numOfTables:%d not 0", pStable->info.tableId, pStable->numOfTables);
     mgmtSendSimpleResp(pMsg->thandle, TSDB_CODE_OTHERS);
   } else {
-    SSdbOperDesc oper = {
+    SSdbOper oper = {
       .type = SDB_OPER_GLOBAL,
       .table = tsSuperTableSdb,
       .pObj = pStable
@@ -806,7 +806,7 @@ static int32_t mgmtAddSuperTableTag(SSuperTableObj *pStable, SSchema schema[], i
   pStable->numOfColumns += ntags;
   pStable->sversion++;
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsSuperTableSdb,
     .pObj = pStable,
@@ -837,7 +837,7 @@ static int32_t mgmtDropSuperTableTag(SSuperTableObj *pStable, char *tagName) {
   int32_t schemaSize = sizeof(SSchema) * (pStable->numOfTags + pStable->numOfColumns);
   pStable->schema = realloc(pStable->schema, schemaSize);
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsSuperTableSdb,
     .pObj = pStable,
@@ -872,7 +872,7 @@ static int32_t mgmtModifySuperTableTagName(SSuperTableObj *pStable, char *oldTag
   SSchema *schema = (SSchema *) (pStable->schema + (pStable->numOfColumns + col) * sizeof(SSchema));
   strncpy(schema->name, newTagName, TSDB_COL_NAME_LEN);
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsSuperTableSdb,
     .pObj = pStable,
@@ -931,7 +931,7 @@ static int32_t mgmtAddSuperTableColumn(SDbObj *pDb, SSuperTableObj *pStable, SSc
     mgmtDecAcctRef(pAcct);
   }
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsSuperTableSdb,
     .pObj = pStable,
@@ -968,7 +968,7 @@ static int32_t mgmtDropSuperTableColumn(SDbObj *pDb, SSuperTableObj *pStable, ch
     mgmtDecAcctRef(pAcct);
   }
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsSuperTableSdb,
     .pObj = pStable,
@@ -1116,7 +1116,7 @@ void mgmtDropAllSuperTables(SDbObj *pDropDb) {
     }
 
     if (strncmp(pDropDb->name, pTable->info.tableId, dbNameLen) == 0) {
-      SSdbOperDesc oper = {
+      SSdbOper oper = {
         .type = SDB_OPER_LOCAL,
         .table = tsSuperTableSdb,
         .pObj = pTable,
@@ -1354,7 +1354,7 @@ static SChildTableObj* mgmtDoCreateChildTable(SCMCreateTableMsg *pCreate, SVgObj
     }
   }
   
-  SSdbOperDesc desc = {0};
+  SSdbOper desc = {0};
   desc.type = SDB_OPER_GLOBAL;
   desc.pObj = pTable;
   desc.table = tsChildTableSdb;
@@ -1508,7 +1508,7 @@ static int32_t mgmtAddNormalTableColumn(SDbObj *pDb, SChildTableObj *pTable, SSc
     mgmtDecAcctRef(pAcct);
   }
  
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsChildTableSdb,
     .pObj = pTable,
@@ -1542,7 +1542,7 @@ static int32_t mgmtDropNormalTableColumn(SDbObj *pDb, SChildTableObj *pTable, ch
     mgmtDecAcctRef(pAcct);
   }
  
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsChildTableSdb,
     .pObj = pTable,
@@ -1687,7 +1687,7 @@ void mgmtDropAllChildTables(SDbObj *pDropDb) {
     }
 
     if (strncmp(pDropDb->name, pTable->info.tableId, dbNameLen) == 0) {
-      SSdbOperDesc oper = {
+      SSdbOper oper = {
         .type = SDB_OPER_LOCAL,
         .table = tsChildTableSdb,
         .pObj = pTable,
@@ -1716,7 +1716,7 @@ static void mgmtDropAllChildTablesInStable(SSuperTableObj *pStable) {
     }
 
     if (pTable->superTable == pStable) {
-      SSdbOperDesc oper = {
+      SSdbOper oper = {
         .type = SDB_OPER_LOCAL,
         .table = tsChildTableSdb,
         .pObj = pTable,
@@ -1805,7 +1805,7 @@ static void mgmtProcessDropChildTableRsp(SRpcMsg *rpcMsg) {
     return;
   }
   
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsChildTableSdb,
     .pObj = pTable
@@ -1848,7 +1848,7 @@ static void mgmtProcessCreateChildTableRsp(SRpcMsg *rpcMsg) {
       mError("table:%s, failed to create in dnode, thandle:%p result:%s", pTable->info.tableId,
              queueMsg->thandle, tstrerror(rpcMsg->code));
       
-      SSdbOperDesc oper = {
+      SSdbOper oper = {
         .type = SDB_OPER_GLOBAL,
         .table = tsChildTableSdb,
         .pObj = pTable
diff --git a/src/mnode/src/mgmtUser.c b/src/mnode/src/mgmtUser.c
index 9098a0c17d4db0c8b955cc5b97c187f1d56262d5..dc4c3d6bea1b1193ec7fb9d05868ce7df23b2051 100644
--- a/src/mnode/src/mgmtUser.c
+++ b/src/mnode/src/mgmtUser.c
@@ -36,12 +36,12 @@ static void mgmtProcessCreateUserMsg(SQueuedMsg *pMsg);
 static void mgmtProcessAlterUserMsg(SQueuedMsg *pMsg);
 static void mgmtProcessDropUserMsg(SQueuedMsg *pMsg);
 
-static int32_t mgmtUserActionDestroy(SSdbOperDesc *pOper) {
+static int32_t mgmtUserActionDestroy(SSdbOper *pOper) {
   tfree(pOper->pObj);
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtUserActionInsert(SSdbOperDesc *pOper) {
+static int32_t mgmtUserActionInsert(SSdbOper *pOper) {
   SUserObj *pUser = pOper->pObj;
   SAcctObj *pAcct = mgmtGetAcct(pUser->acct);
 
@@ -56,7 +56,7 @@ static int32_t mgmtUserActionInsert(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtUserActionDelete(SSdbOperDesc *pOper) {
+static int32_t mgmtUserActionDelete(SSdbOper *pOper) {
   SUserObj *pUser = pOper->pObj;
   SAcctObj *pAcct = mgmtGetAcct(pUser->acct);
 
@@ -67,7 +67,7 @@ static int32_t mgmtUserActionDelete(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtUserActionUpdate(SSdbOperDesc *pOper) {
+static int32_t mgmtUserActionUpdate(SSdbOper *pOper) {
   SUserObj *pUser = pOper->pObj;
   SUserObj *pSaved = mgmtGetUser(pUser->user);
   if (pUser != pSaved) {
@@ -77,14 +77,14 @@ static int32_t mgmtUserActionUpdate(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtUserActionEncode(SSdbOperDesc *pOper) {
+static int32_t mgmtUserActionEncode(SSdbOper *pOper) {
   SUserObj *pUser = pOper->pObj;
   memcpy(pOper->rowData, pUser, tsUserUpdateSize);
   pOper->rowSize = tsUserUpdateSize;
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtUserActionDecode(SSdbOperDesc *pOper) {
+static int32_t mgmtUserActionDecode(SSdbOper *pOper) {
   SUserObj *pUser = (SUserObj *) calloc(1, sizeof(SUserObj));
   if (pUser == NULL) return TSDB_CODE_SERV_OUT_OF_MEMORY;
 
@@ -137,7 +137,7 @@ int32_t mgmtInitUsers() {
   mgmtAddShellShowMetaHandle(TSDB_MGMT_TABLE_USER, mgmtGetUserMeta);
   mgmtAddShellShowRetrieveHandle(TSDB_MGMT_TABLE_USER, mgmtRetrieveUsers);
   
-  mTrace("user data is initialized");
+  mTrace("table:users table is created");
   return 0;
 }
 
@@ -154,7 +154,7 @@ void mgmtReleaseUser(SUserObj *pUser) {
 }
 
 static int32_t mgmtUpdateUser(SUserObj *pUser) {
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsUserSdb,
     .pObj = pUser,
@@ -202,7 +202,7 @@ int32_t mgmtCreateUser(SAcctObj *pAcct, char *name, char *pass) {
     pUser->superAuth = 1;
   }
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsUserSdb,
     .pObj = pUser,
@@ -219,7 +219,7 @@ int32_t mgmtCreateUser(SAcctObj *pAcct, char *name, char *pass) {
 }
 
 static int32_t mgmtDropUser(SUserObj *pUser) {
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsUserSdb,
     .pObj = pUser
@@ -493,7 +493,7 @@ void  mgmtDropAllUsers(SAcctObj *pAcct)  {
     if (pUser == NULL) break;
 
     if (strncmp(pUser->acct, pAcct->user, acctNameLen) == 0) {
-      SSdbOperDesc oper = {
+      SSdbOper oper = {
         .type = SDB_OPER_LOCAL,
         .table = tsUserSdb,
         .pObj = pUser,
diff --git a/src/mnode/src/mgmtVgroup.c b/src/mnode/src/mgmtVgroup.c
index 9be95f4087aba02e4394793e9e7a1e4c24040ca4..87d8ddb2b4da298c57f0a369fe5c9e3f69aea692 100644
--- a/src/mnode/src/mgmtVgroup.c
+++ b/src/mnode/src/mgmtVgroup.c
@@ -22,7 +22,7 @@
 #include "tidpool.h"
 #include "tsync.h"
 #include "ttime.h"
-#include "treplica.h"
+#include "tbalance.h"
 #include "mgmtDef.h"
 #include "mgmtLog.h"
 #include "mgmtDb.h"
@@ -48,7 +48,7 @@ static void    mgmtProcessVnodeCfgMsg(SRpcMsg *rpcMsg) ;
 static void mgmtSendDropVgroupMsg(SVgObj *pVgroup, void *ahandle);
 static void mgmtSendCreateVgroupMsg(SVgObj *pVgroup, void *ahandle);
 
-static int32_t mgmtVgroupActionDestroy(SSdbOperDesc *pOper) {
+static int32_t mgmtVgroupActionDestroy(SSdbOper *pOper) {
   SVgObj *pVgroup = pOper->pObj;
   if (pVgroup->idPool) {
     taosIdPoolCleanUp(pVgroup->idPool);
@@ -62,7 +62,7 @@ static int32_t mgmtVgroupActionDestroy(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtVgroupActionInsert(SSdbOperDesc *pOper) {
+static int32_t mgmtVgroupActionInsert(SSdbOper *pOper) {
   SVgObj *pVgroup = pOper->pObj;
   SDbObj *pDb = mgmtGetDb(pVgroup->dbName);
   if (pDb == NULL) {
@@ -104,7 +104,7 @@ static int32_t mgmtVgroupActionInsert(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtVgroupActionDelete(SSdbOperDesc *pOper) {
+static int32_t mgmtVgroupActionDelete(SSdbOper *pOper) {
   SVgObj *pVgroup = pOper->pObj;
   
   if (pVgroup->pDb != NULL) {
@@ -124,7 +124,7 @@ static int32_t mgmtVgroupActionDelete(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtVgroupActionUpdate(SSdbOperDesc *pOper) {
+static int32_t mgmtVgroupActionUpdate(SSdbOper *pOper) {
   SVgObj *pNew = pOper->pObj;
   SVgObj *pVgroup = mgmtGetVgroup(pNew->vgId);
   if (pVgroup != pNew) {
@@ -147,14 +147,14 @@ static int32_t mgmtVgroupActionUpdate(SSdbOperDesc *pOper) {
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtVgroupActionEncode(SSdbOperDesc *pOper) {
+static int32_t mgmtVgroupActionEncode(SSdbOper *pOper) {
   SVgObj *pVgroup = pOper->pObj;
   memcpy(pOper->rowData, pVgroup, tsVgUpdateSize);
   pOper->rowSize = tsVgUpdateSize;
   return TSDB_CODE_SUCCESS;
 }
 
-static int32_t mgmtVgroupActionDecode(SSdbOperDesc *pOper) {
+static int32_t mgmtVgroupActionDecode(SSdbOper *pOper) {
   SVgObj *pVgroup = (SVgObj *) calloc(1, sizeof(SVgObj));
   if (pVgroup == NULL) return TSDB_CODE_SERV_OUT_OF_MEMORY;
 
@@ -199,7 +199,7 @@ int32_t mgmtInitVgroups() {
   mgmtAddDClientRspHandle(TSDB_MSG_TYPE_MD_DROP_VNODE_RSP, mgmtProcessDropVnodeRsp);
   mgmtAddDServerMsgHandle(TSDB_MSG_TYPE_DM_CONFIG_VNODE, mgmtProcessVnodeCfgMsg);
 
-  mTrace("vgroup is initialized");
+  mTrace("table:vgroups is created");
   
   return 0;
 }
@@ -213,7 +213,7 @@ SVgObj *mgmtGetVgroup(int32_t vgId) {
 }
 
 void mgmtUpdateVgroup(SVgObj *pVgroup) {
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsVgroupSdb,
     .pObj = pVgroup,
@@ -249,7 +249,7 @@ void mgmtCreateVgroup(SQueuedMsg *pMsg, SDbObj *pDb) {
   strcpy(pVgroup->dbName, pDb->name);
   pVgroup->numOfVnodes = pDb->cfg.replications;
   pVgroup->createdTime = taosGetTimestampMs();
-  if (replicaAllocVnodes(pVgroup) != 0) {
+  if (balanceAllocVnodes(pVgroup) != 0) {
     mError("db:%s, no enough dnode to alloc %d vnodes to vgroup", pDb->name, pVgroup->numOfVnodes);
     free(pVgroup);
     mgmtSendSimpleResp(pMsg->thandle, TSDB_CODE_NO_ENOUGH_DNODES);
@@ -257,7 +257,7 @@ void mgmtCreateVgroup(SQueuedMsg *pMsg, SDbObj *pDb) {
     return;
   }
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsVgroupSdb,
     .pObj = pVgroup,
@@ -289,7 +289,7 @@ void mgmtDropVgroup(SVgObj *pVgroup, void *ahandle) {
   } else {
     mTrace("vgroup:%d, replica:%d is deleting from sdb", pVgroup->vgId, pVgroup->numOfVnodes);
     mgmtSendDropVgroupMsg(pVgroup, NULL);
-    SSdbOperDesc oper = {
+    SSdbOper oper = {
       .type = SDB_OPER_GLOBAL,
       .table = tsVgroupSdb,
       .pObj = pVgroup
@@ -596,7 +596,7 @@ static void mgmtProcessCreateVnodeRsp(SRpcMsg *rpcMsg) {
     SQueuedMsg *newMsg = mgmtCloneQueuedMsg(queueMsg);
     mgmtAddToShellQueue(newMsg);
   } else {
-    SSdbOperDesc oper = {
+    SSdbOper oper = {
       .type = SDB_OPER_GLOBAL,
       .table = tsVgroupSdb,
       .pObj = pVgroup
@@ -659,7 +659,7 @@ static void mgmtProcessDropVnodeRsp(SRpcMsg *rpcMsg) {
 
   if (queueMsg->received != queueMsg->expected) return;
 
-  SSdbOperDesc oper = {
+  SSdbOper oper = {
     .type = SDB_OPER_GLOBAL,
     .table = tsVgroupSdb,
     .pObj = pVgroup
@@ -716,7 +716,7 @@ void mgmtDropAllVgroups(SDbObj *pDropDb) {
     if (pVgroup == NULL) break;
 
     if (strncmp(pDropDb->name, pVgroup->dbName, dbNameLen) == 0) {
-      SSdbOperDesc oper = {
+      SSdbOper oper = {
         .type = SDB_OPER_LOCAL,
         .table = tsVgroupSdb,
         .pObj = pVgroup,
diff --git a/src/vnode/src/vnodeMain.c b/src/vnode/src/vnodeMain.c
index 393758761d2fa63c22eaa0d93350b1f2ff58b661..84ae662110e59b8f43f496bb993e4194f5a68aed 100644
--- a/src/vnode/src/vnodeMain.c
+++ b/src/vnode/src/vnodeMain.c
@@ -45,6 +45,9 @@ static pthread_once_t  vnodeModuleInit = PTHREAD_ONCE_INIT;
 #ifndef _SYNC
 tsync_h syncStart(const SSyncInfo *info) { return NULL; }
 int     syncForwardToPeer(tsync_h shandle, void *pHead, void *mhandle) { return 0; }
+void    syncStop(tsync_h shandle) {}
+int     syncReconfig(tsync_h shandle, const SSyncCfg * cfg) { return 0; }
+int     syncGetNodesRole(tsync_h shandle, SNodesRole * cfg) { return 0; }
 #endif
 
 static void vnodeInit() {