Kristina Chodorow (@kchodorow) explains MongoDB’s sharding implementation (nb still in alpha at the time of this writing):
Sharding is the not-so-revolutionary way that MongoDB scales writes (it’s very similar to techniques described in the Big Table paper and by PNUTS)
Credit Kristina Chodorow
On the same topic of MongoDB sharding, you’ve probably heard already about the “8000000 operations per second with our sharded storage” published on the same blog. I’d like to strongly encourage you once again to check the facts (nb which are
almost completely missing this time) before admitting such benchmarks/numbers.