I haven’t used R recently (10 years or so), but when I did, the speed with which K/kdb+ could scan through and summarize terabytes of data was orders of magnitu

I haven’t used R recently (10 years or so), but when I did, the speed with which K/kdb+ could scan through and summarize terabytes of data was orders of magnitude faster than R or any other system. Once the data was summarized into (say) a gigabyte or so, analyzing it with R or even Python was much easier thanks to the ecosystem and reasonable time (probably 10-100 times slower, but the time saved by using well tested stat code is more than worth it)