With the constant increase of networks in Companies, the crucial necessity of a high availability and performance is undeniable. The decentralization of the traffic, for multiples servers (for example), will facilitate the attendance for more requisitions beyond improving importantly the performance of the utilized service.
The Load Balance can be utilized in several scenarios. As well as, for load distribution in servers or in network devices. Therefore, avoiding that only one component stays responsible for meeting all the requisitions.
SCENARIO
Imagine a famous e-commerce site. Now, try to figure the damage that this shop would have if their Website crashes. Hence, in a pick of accesses, probably, only one server will not be able to handle it. With the usage of the load balancing, multiple servers will share the workload preventing consequently its overloading.
With the load balancing applied, besides the gain in performance, the service will have an increase in reliability. As well as, tolerance to failures (in case there are problems with a server, the traffic will be diverted for the others). In a world progressively more connected, the “uptime” of your service is increasingly more fundamental.
Now that we understand the Load Balancing and its benefits. How will we do the monitoring in the TRAFip?
MONITORING OF THE LOAD BALANCE IN TRAFIP
For this example, we will use the scenery of Load Balancing between servers.
In the first moment, it will be important to register every server on the TRAFip as a subnetwork. After registering, it will be necessary to create a group of subnetworks with all the servers participating in the Load Balancing.
With the scenery prepared it will be time for creating the traffic profile. Since we want to analyze the participation of each member in the total traffic of the group that himself(herself) is a part of, we will also need the TRAFip to make an analysis in the same direction of the traffic. In other words, when observing the traffic originated in a certain group, we need also to be able to specify the server that originated the traffic, and when observing the destiny traffic of the group, specifying to which server the traffic has been destined to.
Did you realize? In the traffic of origin, the IP of origin will be analyzed and in the traffic of destiny, the IP of destiny will also be reviewed. The analysis is always done in the same way of the traffic. This is the main characteristic of the profile in distribution.
In curves of the traffic we will select the subnetworks that are members of the group. Subnetworks which represent each one of the servers.
Finally, we have applied the profile in distribution, in the group of subnetworks, to which the servers are members. Now, it will be enough to navigate until the graffic of the group and click on the profile which will appear above.
With the applied profile, we are capable of observing the distribution of traffic between the servers, such as we see in the graphics above. Therefore, guaranteeing that the strategy defined for the distribution of the traffic is actually being obeyed, in practice. Finally, having the possession of the data for validating and accompanying the functioning of the load balancing.
FINAL CONSIDERATIONS
TRAFip is a system of traffic analysis which allows the determining of what, how, when, where and by whom your network is being utilized.
In this way, there are no doubts about the importance of investing on network management. In this same manner, bringing not only benefits to the network visibility but also being a complementary way to seek for the prevention of problems that might cause the dropping of network services.
Thinking of that, Telcomanager present in the market since 2002, and a leading Latin America brand in the sector of software for managing networks. Also counting with a unique and innovative technology, deploying smart solutions in the monitoring of data that will provide a stratified vision of the traffic, is now allowing your Company to follow the most important aspects of your network, in real time.