High availability and load balancing for web services
Kumpunen, Jukka (2019)
Kumpunen, Jukka
2019
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-201904165438
https://urn.fi/URN:NBN:fi:amk-201904165438
Tiivistelmä
The goal of this thesis project was to get familiar with high availability and load balancing for servers and websites. This was done by studying the benefits that high availability and load balancing offer for companies, and why using them is important especially for large company websites. The purpose of the project was to provide information for companies whose service unavailability can cause financial loss or affect customer satisfaction. With the help of the methods covered in this thesis, these types of situations can be avoided.
The methods companies use to implement high availability and load balancing vary from company to company. There are many viable solutions, such as software- and hardware-based load balancing and load balancing using cloud services or the domain name system (DNS). The price of high availability and technical difficulty of administration depend on the chosen method and the number of users a service has.
Companies must target a certain level of error tolerance that depends on the services they offer. Critical services whose unavailability quickly affects customers and the revenue of a company must be implemented in a way that can tolerate server issues and significant, unexpected changes in popularity. Methods suitable for this include software-based high-availability and cloud-based load-balancing solutions. The results of this project can be used by companies that are planning the server infrastructure and administration of their upcoming online services.
The methods companies use to implement high availability and load balancing vary from company to company. There are many viable solutions, such as software- and hardware-based load balancing and load balancing using cloud services or the domain name system (DNS). The price of high availability and technical difficulty of administration depend on the chosen method and the number of users a service has.
Companies must target a certain level of error tolerance that depends on the services they offer. Critical services whose unavailability quickly affects customers and the revenue of a company must be implemented in a way that can tolerate server issues and significant, unexpected changes in popularity. Methods suitable for this include software-based high-availability and cloud-based load-balancing solutions. The results of this project can be used by companies that are planning the server infrastructure and administration of their upcoming online services.