Hello everyone welcome to Malik Jaya 34 channel this is Malik Arjun today I'm going to explain about the types of databases in Oracle database there are four basically there are four types of Oracle databases we can we will usually see in Oracle database types for example a standalone standalone database with Oracle Restart as a service RAC, One Node and RAC database. So I am going to explain what are the characteristic difference of all these three database. Before I get started with today's session, if you are new to my channel, I request you to go back to YouTube and search for malik034 and then please do subscribe.
And if you missed my previous videos. always go back to youtube and search for malik034 and then please do watch all those videos if you like my videos please share like and comment let's start with the today's session so i said standalone it's a normal traditional rdbms and standalone with oracle restart service is a again you know database with asm as a standalone database and rack one node is a one instance node with in the cluster database and rack database is a cluster database if you see stand a standalone database it will be like the storage will be our local Linux or local storage file system like local mount point I can say or NFS you can have or you know SAN you can have or NAS you can have whatever you know mount point it is but it is locally mounted mounted on on that particular server for example what are data archive like a local mount points or remote file system mount points so on that my database will be running that is like a standard standalone database and if you come to oracle restart service or standalone service so you are basically going to create a asm disk group on top of that asm disk group you are creating a data or whatever disk group on top of that you are installing the asm instance as a cluster where grid you know restart service and then on top of that asm you are going to install your database so that is oracle restart so it is not going to be a normal local file system or nfs or remote file setup it is going to be on the disk group and whereas in rack one node it's the same as oracle restart you are going to create asm disk groups and you are going to install asm instance and you are going to create a data recall whatever disk group on top of that you are going to install your database in case of rack it is as usual like all asm disk groups on top of asm disk group you are going to create your database so how many servers you need it in standalone we need only one server in that server we are going to install our database and in case of oracle restart service again we need one server on that server we are going to install asm on top of that we are going to install our database whereas in rack one node The number of servers depends upon number of cluster. If I have a 2 node cluster or 3 node cluster or 4 node cluster On all 4 nodes we are going to install ASM and then on all 4 nodes we are going to create the data and record script and I am going to run my ASM instance on any one of the cluster nodes.
For example, 4 node cluster I have but my ASM will be running on any one node not on all 4 nodes. Whereas in case of RAC Asm will be installed on all four nodes and my database instance will be running on all four nodes The difference between the rack one node and rack node is in the rack all the instance will be running on all the nodes But whereas in rack one node only one instance will be running on any one of the cluster nodes right, so RAC1 node and RAC nodes are clustered databases but the difference is in RAC1 node only one instance is running in RAC all the instances were running at the same time Coming to the third difference Is ASM needed in standalone? We don't need ASM In all other three cases we need ASM so in in oracle restart the asm instance name is going to be just asm whereas rack one node and rack the asm instance is going to be asm instance one two three four five up to how many cluster nodes you have for suppose if you have four cluster node asm1 asm2 asm3 asm4 four asm instances running on all four nodes whereas in rack also if the four node cluster all four node all four asm will be running on all four node asm1 two three four and and number of instances in standalone there will be only one database instance in oracle restart again there is only one instance in rack one node always there will be only one instance running on any one of the cluster nodes whereas in rack it's a n number of instances how many cluster nodes we have those many instances will be running or again it depends how you configure it but in ideally n number of instances in the case of rack so what will be the typical instance name so if prod is my database name in the standalone it is going to be prod db and vertical restart is going to be prod db and in case of Rack1 node the instance name is going to be proddba1, proddba2, proddba3, proddba4 Similarly how many cluster nodes you have those many instances will be running So but at any time only one instance will be running like proddba1 will be running on node1 and other 3 will be down so in case if this proddb underscore one goes down automatically other you know cluster instance will come up like proddb underscore 2 will come up and proddb underscore 2 which goes up and automatically next instance proddb 3 will come up so that is the reason oracle one node is it will be running only one instance on any one of the cluster nodes so if anything goes down you know automatically the other uh cluster instance will be uh you know other node instance will be come up so whereas in case of rack all the instances like prod dp1 prod db2 prod db3 prod db4 if it is a four node rack all four all four node instance will be up and running at the same time that's a rack that is the difference between rack one node and rack so all four instances will be running at the same time here only one instance will be running at at any one time and if that goes down automatically other instance will come up.
So the naming convention is going to be underscore in case of rack1 node and in rack it is going to be just a numeric 123 at the end of the database. so, Oracle auto restart, in case of suppose your server bounce some network issue or some OS issue your database server goes up and then it will come up in case of standalone my database will not start whereas in case of Oracle restart automatically ASM and my database will come up whereas rack one node and rack both same like automatically my ASM will restart so if for example like other differences if the database goes down here you know it will manually we need to restart my database here in this rack Oracle standalone restart service if my database goes down ASM will automatically start my database here also in rack one node and rack database if database goes down asm will automatically try to restart it okay that is the difference in the database so in case of server reboot happens in standalone again manually we need to start my database in case of Oracle standalone in restart service or rack one node or rack my server reboots automatically ASM will come up once the ASM comes up automatically it will start my database so that is the advantages of you know having all these different types of databases. So if you compare with the standalone and Oracle Restart both are functionality same but one added advantage is with the restart services is ASM can manage your instance so automatically it will start and stop whenever there is an issue with the instance and the data will be managed by ASM instance in the data disk group or record disk group so we don't need of any value manager or we don't need of any you know the value functionality value manager functionality like data striping or data accessing or you know data mirroring so everything will be taken care by ASM here everything will be taken care by OS so and manual manual integration is more in the standalone database whereas in rack one node and rack both the functionalities are same and both are managed by ASM and both data is resized on data disk groups and both are mirrored and striped data across all the disk groups and ASM is going to manage my database and whenever server reboots or database reboots ASM will take care of restarting my database the only advantage is in rack onenode will be having so much throughput because number of instances will be running at the same time the number of connected will be more established here whereas in case of rack one node so only one instance will be running so the there is a limitation in the number of connections and uh you know in case of that node goes down automatically uh you know other instances will come up comes up so that's the only difference between the rack one node and rack database so if i go to next slide so suppose like i have four so if you have four nodes and if you want to create a standalone database so you can create a dev db on one node and test db and other node and sit db another node and prod db another node you can have four environment like you have to install your oracle homes db of one and this node one db home two on this node two db home three and this node three and db home four on this node four and you can have four environment ready and then you can use it but there is no high availability there is no load balancing and if this database goes down or server goes down you will lose your entire enrollment right so that's the difference in the standalone and whereas in oracle restart service so oracle restart service also what happens first you need to install ASM on this SM SM SM SM so there is no diff there is no connectivity between all these three SM these all ASM whatever you are going to install like SM SM SM SM like you can install a SM on node 1 yes ASM1 node 2, ASM1 node 3, ASM1 node 4, all 4 ASMs are individual to each other. So, there is no common communication between both of all these 4 nodes. So, these are individual environment.
So, like that you can install griddom. For this ASM, you can install griddom 1, griddom 2, griddom 3, griddom 4. You can install all these griddom 4, all griddom 1, 2, 3, 4 and all these 4 nodes. On top of that ASM, you can install your devdb, testdb, sitdb and proddb. And again, 4 nodes.
4 environment you can create with db-home1 db-home2 and db-home3 and db-home4 like you can install 4 db-home individually and 4 grid-home individually and 4 asm instance individually and 4 database like dev test sit and prod individually and you can have 4 environment ready and then you can use it the the advantage here is if this node goes down or if this database goes down asm will restart it automatically so that is the only difference between the standalone and oracle restart whereas oracle one node what you are going to do here if you quickly observe here you are going to install SM1 here you can see SM1 SM2 SM3 SM4 you are going to install SM1 SM2 SM3 and SM4 but that grid dome whatever you are installing here user app oracle grid user an app oracle grid user an app oracle grid whatever you are going to install your grid infrastructure home here is it's a shade it is a common across common in the sense the it is a cluster grid home you are going to install all and it is act as a one grid home combination of all these fours it is not an individual grid home so and then ASM instance whatever you are creating is managed by that particular grid home ASM 1 ASM 2 ASM 3 ASM 4 right so then you can install your database on all those disk groups whatever created across this disks uh data disk or record disk so the database instance is going to be dev db underscore one dev db underscore two dev db underscore three dvd underscore four dvd underscore one dvd underscore two dvd underscore three dvd underscore four but at any single point of time only one instance will be running on any one of the node if dev DB underscore one is running here if this goes down automatically dev DB underscore two will come up so that is the advantage of having oracle one node so automatically the instance will be started on the other node if this server goes down automatically instance will be shifted to node 3 or node 4 right based upon the policy right so where is in case of rack you are going to install the same grid infrastructure home on all four db nodes sm1 sm2 sm3 sm4 and then you are going to run your devdb1 devdb2 devdb3 devdb4 all four instances on all four nodes at the same time. So that is the difference between the RAC1 node and RAC database. So if you are having the RAC1 node or RAC database, you are going to achieve only one instance.
If you see here one enrollment. I am going to achieve only Dev DB as my environment here whereas in case of stand alone and oracle restart I am going to achieve Dev DB test DB SIT DB and Pro DB four environment individually and whereas rack one node and rack like all four node together I am going to achieve only one environment right so that is the difference having uh you know all these three four types of databases each has its own advantages and each has its own disadvantages but ideally widely used uh you know suggested or more robust and more uh you know band more throughput and high availability load balancing so one of the best recommendation is always go with the rack database right you can do dynamic scaling and dynamic scale out and scaling everything is possible in case of rack so now that's the best suggested way to you know implement in any of the enterprise architect okay that's it guys thank you