12,831 karma · joined March 22, 2007
Along with various other nice work about neural networks:
http://cs.stanford.edu/people/karpathy/convnetjs/started.htm...
http://cs.stanford.edu/people/karpathy/recurrentjs/
http://karpathy.github.io/neuralnets/
Nice visualization about classifiers from the class (various svm, softmax):
http://vision.stanford.edu/teaching/cs231n/linear-classify-d...
'Hong Kong’s bourse doesn’t allow share classes with different voting rights, as the U.S. does. Such arrangements helped Zuckerberg and Google co-founders Larry Page and Sergey Brin keep control of their companies after they went public."
Taking the case of minimally loaded, you need to keep track of how many active requests each node/replica is serving, as well as globally keeping track of the min. (which past a certain load, will suffer a lot of contention to update)
To do choice of 2, all you need is to keep track of active requests per node/replica.
Under spiky workloads, there is also a problem with choosing minimally loaded. The counter for numRequests of a node might not update fast enough, so that a bunch of requests will go to that node, quickly saturating its capacity.
Choice of 2 doesn't suffer this problem bec of its inherent randomization.
I understand that taking a break and getting perspective while in the midst of a lot of work can be effective, but this seems a little bit extravagant.
Sidenote: benbernard worked on Amazon dev tools team