Additional Topics¶

MapReduce programming model, implementation strategies using multiple threads and/or processes, comparison to previous systems that do the same thing, but not as well.
Caching, memory hierarchies, registers, L1, L2 and L3 caches, main memory, disk, swap space, data locality, temporal locality, custom caches, cache replacement policies (LRU, MRU, TLRU, and so forth), profiling tools, callgrind, cachegrind, and alternatives
Non-blocking I/O, where normally slow system calls like open, accept, read, and write return immediately instead of blocking, select, epoll_* set of functions, libev and libuv open source libraries.

Lectures¶