Erdös started taking benzedrine in 1971, long after he had become one of the most prolific mathematicians of all time with ~40 years of work behind him. I suspect he was productive in spite of the speed, as someone that brilliant with that many years of learning in their head would still be incredibly brilliant impaired.
I would love to see this at hackathons and conferences everywhere. So many people need help but are so ashamed to ask for it (particularly in our field) that you literally need to make talking to someone as convenient as getting a McDonald's hamburger to get them to seek help.
It depends completely on the application. It's also usually best to focus on the tail end -- I've always found alerts on 95th or 99th percentile latency useful and easy to decide on thresholds for (ask yourself -- how slow can I tolerate this being for 1% or 5% of users?)
This reminds me a little of the Netezza data warehouse appliance's architecture: a query planner in front of lots of little nodes with one disk, one CPU, and an FPGA. Every query is a full table scan, each node flashes the WHERE clause to the FPGA, and slurps the whole disk through the FPGA.
I believe based on context that you and the article are talking about different kinds of parallelism. Yours is the more usual sense, the author seems to be referring to the fact that CRC does not exhibit instruction level parallelism which out-of-order execution exploits. This is because the loop is very tight (probably a few instructions) and each iteration depends on the results of the previous, so there is only a tiny window for reordering instructions.
This kind of talk is what bothers me about investment in this industry -- I've never seen anywhere else where people are so OK with no ROI. This is where all the dumb money in SV is coming from.
It's like all statistical tests -- it works really well (provably well) when the assumptions it requires hold. However, it's usually impossible to know if those assumptions hold without holding the desired answer in the first place. That's why nonparametric tests are so popular (not saying they have much to do with the article at hand, but people are definitely willing to get less definitive results in exchange for making fewer assumptions).