> In many cases, you’d be better off running the same computation on your laptop.
Stopped reading there. If you're better off running the same computation on your laptop, you're just not dealing with real big data and so you don't need any distributed systems. Simple as that.
That's his whole point. But he goes further, by showing via experiment that a lot of distributed systems are solving problem sizes in benchmarks that can be solved quicker in a single thread.
Stopped reading there. If you're better off running the same computation on your laptop, you're just not dealing with real big data and so you don't need any distributed systems. Simple as that.