Load Balancing and Clustering Question
1) What is difference between Clustering and load Balancing (With out cluster)
Main Object of both are used to achieve the High Availability.
Cluster :
1) Mainly used to connect the group of server which used share the resource between the server
2) Maintains the state between the server ie servers are dependent to each other
Load Balancing
1) Group of servers are connected to LB in Independent manner ie each server acting independently.
2) Since its independent it will not share the resource
User Preference (Cluster / LB):
Let me explain the simple scenario , User try to do the online shopping .Due to some reason one of the server is down.
Cluster Behavior
In the cluster environment user will continue the shopping with out any breaking session ie cluster will redirect to another server with same session .
LB Behavior :
In LB , user needs to starts from scratch since one of the server is down . LB will start the new request and redirect to another server.
2) What are Types in Cluster ?
Two ways to implement the cluster to achieve the HA .
1) Apps Server Cluster
2) DB Server Cluster
3) What are the advantages of LB?
1)To achieve the HA
2) To achieve the High through put
3) Minimize the Response time
3) What are Techniques in LB ?
1) Round Robin
2) Session Affinity/Sticky Session
3) IP Request Affinity
3) What is Fail Over and How do we achieve the Fail Over ?
Fail over to achieve the HA , if any one machine down it automatically stop serving request to that particular server instead it will redirect another server
In LB Level , We have a feature called Heart beat check , once LB does not receive any Heart Beat from server then it assume that it will down and stop serving the request to that particular server.
4)How do we handle the session failover in the apps server?
Through Session Replication we can handle the session Fail over ie session values are replicated to all the server in cluster env.
In this web application we can enable the session fail over and load balancing through tag /attribute in web .xml
In this web application , Session replication should be enable not only web.xml in addition all the attributes /object should be serilizable then only it will replicated through cluster
5) Way of session replication
File based Session replication
DB Based Session Replication
Memory Based Session Replication
6) What is HA?
High availability means ability to access the system with out any interrupt.
HA with low latency and with High Through put is main objective of all the business
7) What is scalablity ?
To handle the amount of load to increase the system resources either scale up or scaled out .
Scale up :Increase the resources in existing node ie increase RAM/Processor or etc
Scaled Out : Increase/add the additional nodes in existing env.
Both way to impact the cost and time line of the projects
Main Object of both are used to achieve the High Availability.
Cluster :
1) Mainly used to connect the group of server which used share the resource between the server
2) Maintains the state between the server ie servers are dependent to each other
Load Balancing
1) Group of servers are connected to LB in Independent manner ie each server acting independently.
2) Since its independent it will not share the resource
User Preference (Cluster / LB):
Let me explain the simple scenario , User try to do the online shopping .Due to some reason one of the server is down.
Cluster Behavior
In the cluster environment user will continue the shopping with out any breaking session ie cluster will redirect to another server with same session .
LB Behavior :
In LB , user needs to starts from scratch since one of the server is down . LB will start the new request and redirect to another server.
2) What are Types in Cluster ?
Two ways to implement the cluster to achieve the HA .
1) Apps Server Cluster
2) DB Server Cluster
3) What are the advantages of LB?
1)To achieve the HA
2) To achieve the High through put
3) Minimize the Response time
3) What are Techniques in LB ?
1) Round Robin
2) Session Affinity/Sticky Session
3) IP Request Affinity
3) What is Fail Over and How do we achieve the Fail Over ?
Fail over to achieve the HA , if any one machine down it automatically stop serving request to that particular server instead it will redirect another server
In LB Level , We have a feature called Heart beat check , once LB does not receive any Heart Beat from server then it assume that it will down and stop serving the request to that particular server.
4)How do we handle the session failover in the apps server?
Through Session Replication we can handle the session Fail over ie session values are replicated to all the server in cluster env.
In this web application we can enable the session fail over and load balancing through tag /attribute in web .xml
In this web application , Session replication should be enable not only web.xml in addition all the attributes /object should be serilizable then only it will replicated through cluster
5) Way of session replication
File based Session replication
DB Based Session Replication
Memory Based Session Replication
6) What is HA?
High availability means ability to access the system with out any interrupt.
HA with low latency and with High Through put is main objective of all the business
7) What is scalablity ?
To handle the amount of load to increase the system resources either scale up or scaled out .
Scale up :Increase the resources in existing node ie increase RAM/Processor or etc
Scaled Out : Increase/add the additional nodes in existing env.
Both way to impact the cost and time line of the projects
Comments
Post a Comment