What are Different Types of Load Balancers?
In computing, load-leveling refers to the method of distributing a collection of tasks over a collection of resources, with the aim of creating their overall process a lot more economical. A load balancing software continually tries to answer a particular downside.
Among alternative things, the character of the tasks, the recursive complexity, and the hardware design on that the algorithms can run furthermore error tolerance, should be taken into consideration. Thus, compromise should be found to best meet application-specific necessities
Elastic Load leveling supports the following sorts of load Balancers:
Application Load Balancers, Network Load Balancers, and Classic Load Balancers. Amazon ECS services will use either variety of load balancers.
Application Load Balancers square measure accustomed route HTTP/HTTPS (or Layer 7) traffic. Network Load Balancers and Classic Load Balancers square measure accustomed route protocol (or Layer 4) traffic.
Application load balancers
An Application Load Balancer makes routing choices at the appliance layer (HTTP/HTTPS), supports path-based routing, and might route requests to at least one or a lot of ports on every instrumentation instance in your cluster.
Application Load Balancers support dynamic host port mapping. as an example, if your task’s instrumentation definition specifies port eighty for AN NGINX instrumentation port and port zero for the host port, then the host port is dynamically chosen from the temporary port vary of the instrumentation instance (such as 32768 to 61000 on the most recent Amazon ECS-optimized AMI).
Once the task is launched, the NGINX instrumentation is registered with the appliance Load Balancer as AN instance ID and port combination, and traffic is distributed to the instance ID and port equivalent to that instrumentation. This dynamic mapping permits you to possess multiple tasks from one service on a similar instrumentation instance
Network load balancer
A Network Load Balancer makes routing choices at the transport layer. It will handle ample requests per second. Once the load balancer receives an association, it selects a target from the target cluster for the default rule employing a flow hash routing algorithmic program.
It makes an attempt to open a TCP association to the chosen target on the port laid out in the auditor configuration. It forwards the request while not modifying the headers. Network Load Balancers support dynamic host port mapping.
For instance, if your task’s instrumentation definition specifies port eighty for associate NGINX instrumentation port and port zero for the host port, then the host port is dynamically chosen from the temporary port vary of the instrumentation instance (such as 32768 to 61000 on the most recent Amazon ECS-optimized AMI).
Once the task is launched, the NGINX instrumentation is registered with the Network Load Balancer as associate instance ID and port combination, and traffic is distributed to the instance ID and port reminiscent of that instrumentation.
This dynamic mapping permits you to possess multiple tasks from one service on a similar instrumentation instance.
Classic load balancer
A Classic Load Balancer makes routing choices at either the transport layer (TCP/SSL) or the appliance layer (HTTP/HTTPS). Classic Load Balancers presently need a hard and fast relationship between the load balancer port and therefore the instrumentality instance port.
For instance, it’s attainable to map the load balancer port eighty to the instrumentality instance port 3030 and therefore the load balancer port 4040 to the instrumentality instance port 4040.
However, it’s impossible to map the load balancer port eighty to port 3030 on one instrumentality instance and port 4040 on another instrumentality instance. This static mapping needs that your cluster has a minimum of several instrumentality instances because the desired count of one service uses a Classic Load Balancer.
The potency of load reconciliation algorithms critically depends on the character of the tasks. Therefore, the lot of data regarding the tasks is on the market at the time of decision-making, the larger the potential for optimization.
In simple terms, the classic load balancer can be seen as a connection-based balancer, which just forwards the requests from the load balancer to the backend section, without looking into any of the requests.
Bottom-Line: Why is it essential?
It is essential because it helps with service availability. If load balancing is not used, there are chances that a single server could break down and restrict the availability.
If a server is not capable of handling requests at an optimum level, response time would be a big fail and customer dissatisfaction would follow the same.
Another reason is that load balancing is the underlying element for cloud environments. When load balancing software is used scalability can be increased drastically and also helps in solving a huge issue which is overloading.
