Zakeke was designed and built with the goal of scalability as a prior aim.
First of all, our system is fully deployed on the Microsoft Azure cloud platform. Services scale automatically depending on the workload, no supervision is needed and everything is automatically adjusted according to the various parameters of CPU, memory, etc. which identify any workload peaks.
The architecture also uses CDN systems for file distribution, a centralized memory cache system distributed over several processing nodes, a data storage system based on a multi-partition NoSQL architecture, and with throughput automatically modulated based on the level of scaling-out required to cope with the variable load.
If you still have doubts, drop us a line and we would love to tell you more about Zakeke architecture :)