Large and distributed web systems
Web-sites that serve billions of people all over the world, handle millions of transactions per second, can be called large web systems. Examples are Google, Amazon, Yandex, Wikipedia, Netflix, Salesforce etc.
Large systems have to scale. They have to work without downtime and be responsive even during peak-hours. Also, those systems have to be designed to handle quick growth of number of users, which is quite often scenario for some types of websites.
One of the most important feature for the most of large web systems is reliability. They have to be available 24x7x365. All maintenance and fixes have to be done seamless for end users.
Large systems are designed differently from normal applications. Before the development has begun, during the design phase, load, scaleability, reliability, monitoring, infrastructure and various other things have to be thought of. Mistakes conducted here can be fatal for the project.
Another bug task is to support those systems.
How do we prevent outages? How do we update the system without visible downtime? How do we optimize the load? Those are just a few problems that must be solved.
Our team has a vast experience in design, development and support of such systems. We love to share that information with our partners and internet community.