back to article Sparkling new Spark distribution spurs MapR to reduce MapReduce

And on the sixth day (of June) MapR announced its new streaming distribution would use Spark and not MapReduce – though this will complement rather than replace its Hadoop distro. Although San Jose-based MapR tips its hat to MapReduce by name, the increasing obsolescence of Google's 2004 framework – and the public enthusiasm …

  1. Steve Channell

    back to the future

    Who'd have thought that the replacement for Hadoop would be so much like what we used to call a DataBase Management System with an SQL query language, at this rate uniqueness and indexes will come back to in just a few short years.

    1. Adam 52 Silver badge

      Re: back to the future

      Secondary indexes are already there in Apache Phoenix! It's been normal to use sorted column stores (otherwise known as clustered indexes) in Hadoop land for years now.

      Seriously; I'd be very interested in anyone who's made Spark work without having to buy/rent eye watering amounts of hardware that would make a more conventional database fly.

      1. Anonymous Coward
        Anonymous Coward

        Re: back to the future

        Phoenix is a joke.

        1. Adam 52 Silver badge

          Re: back to the future

          That's a shame, looks perfect from the outside. What issues does it have?

          1. Ian Michael Gumby Silver badge

            Re: back to the future

            It's slow, doesn't work well enough and the local indexes are DOA from the start.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2019