Sample Question #290 (programming – algorithms)
We have a huge database that rivals the one at Google in size. How do you propose we search through this database for specific data items in real time? For example, if a client wants to see the last three trades in Google stock on April 1, 2008, what kind of search algorithm would you propose? (The client, along with our 300,000 other clients, wants the answer in less than a second.) Assume we don’t have the million-PC server farm that Google does; in fact, we just have one server running an Intel quad-core CPU.
[This is a very practical case question, although, frankly, if I knew the answer, I’d have tried to find a job at Google (or one of its competitors)]