3 ms·

For perspective, twice Kentucky's entire state population commutes into Manhattan every weekday. Scale matters.

• zerohm 13y ago
  Where are you getting these numbers? I get... Manhattan commuters = 1.6M Kentucky population = 4.38M [1] http://cityroom.blogs.nytimes.com/2013/03/05/more-people-commute-to-manhattan-than-any-other-county/?_r=0 http://cityroom.blogs.nytimes.com/2013/03/05/more-people-com... [2] http://quickfacts.census.gov/qfd/states/21000.html http://quickfacts.census.gov/qfd/states/21000.html
  • 001sky 13y ago
    http://wagner.nyu.edu/files/rudincenter/ManhattanCommuting.pdf http://wagner.nyu.edu/files/rudincenter/ManhattanCommuting.p...
• Dylan16807 13y ago
  The users don't have to interact. If you can serve the first million people without problems, you can scale linearly to 100 million.
  • hga 13y ago
    That ignores the back ends to the back end. E.g. an IRS system that might be able to handle a million people with query/response transactions would almost certainly need different strategies to scale up 100 times.
    • Dylan16807 13y ago
      >That ignores the back ends to the back end. E.g. an IRS system Just make sure your component won't fail. Perhaps offer other departments help with sharding, be that consulting or paying for servers. Give a friendly error message that the other system has gone down, but your data is saved and will be resumed automatically upon next visit. Maybe retry later and email people when it succeeds. >an IRS system that might be able to handle a million people with query/response transactions would almost certainly need different strategies to scale up 100 times. I don't see why it would. Sort by name or SSN and do a roughly-even split across your servers.