Comments on: Twitter to open source Hadoop-like tool http://gigaom.com/2011/08/04/twitter-to-open-source-hadoop-like-tool/ Wed, 19 Jun 2013 13:21:19 +0000 hourly 1 http://wordpress.com/ By: Anh-Tuan GAI http://gigaom.com/2011/08/04/twitter-to-open-source-hadoop-like-tool/#comment-645242 Sun, 07 Aug 2011 08:42:21 +0000 http://gigaom.com/?p=388794#comment-645242 there is no similarity but maybe the first one rely on the other. take a look at http://www.slideshare.net/nathanmarz/the-secrets-of-building-realtime-big-data-systems

]]> By: Daniel Patterson http://gigaom.com/2011/08/04/twitter-to-open-source-hadoop-like-tool/#comment-645206 Sat, 06 Aug 2011 23:33:02 +0000 http://gigaom.com/?p=388794#comment-645206 He said it wouldn’t work on TRILLIONS, and would work on BILLIONS. This is the answer to your question – it will work in real time if your data sets are on the order of billions, but won’t be fast enough to handle trillions.

]]> By: Himanshu Bansal http://gigaom.com/2011/08/04/twitter-to-open-source-hadoop-like-tool/#comment-645167 Sat, 06 Aug 2011 16:04:01 +0000 http://gigaom.com/?p=388794#comment-645167 I am confused among “will make it likely very slow for processing trillions of words.” and “Storm would be very good for continuous real time stats on say billions of words”. Will Storm be good for real time processing or not?

]]> By: Stacey Higginbotham http://gigaom.com/2011/08/04/twitter-to-open-source-hadoop-like-tool/#comment-645048 Fri, 05 Aug 2011 22:33:41 +0000 http://gigaom.com/?p=388794#comment-645048 That’s a bad job on my part with modifiers and punctuation. There should be a comma between Hadoop-like and real-time. As I explain in the post Hadoop is batch and Storm is streamed data.

]]> By: Joseph Heck http://gigaom.com/2011/08/04/twitter-to-open-source-hadoop-like-tool/#comment-644954 Fri, 05 Aug 2011 15:49:50 +0000 http://gigaom.com/?p=388794#comment-644954 Why would you question if another tool is good for the environment? It’s not a direct replica of any of the others, and this technology space is still incredibly immature. The near-real-time stream processing space, in particular, is quite narrow in terms of options.

]]> By: Shahid Ibraheem http://gigaom.com/2011/08/04/twitter-to-open-source-hadoop-like-tool/#comment-644916 Fri, 05 Aug 2011 10:52:47 +0000 http://gigaom.com/?p=388794#comment-644916 http://www.easy-paisa.com/affiliates/uid/shahidali786_1

]]> By: Irfan Habib http://gigaom.com/2011/08/04/twitter-to-open-source-hadoop-like-tool/#comment-644903 Fri, 05 Aug 2011 08:33:53 +0000 http://gigaom.com/?p=388794#comment-644903 ” Hadoop-like real-time data processing tool”

Hadoop is specifically designed for batch processing and not real time processing.

]]> By: Anand Prakash http://gigaom.com/2011/08/04/twitter-to-open-source-hadoop-like-tool/#comment-644853 Fri, 05 Aug 2011 01:17:10 +0000 http://gigaom.com/?p=388794#comment-644853 There is no similarity between Storm and Hadoop. Hadoop is a batch processing system and this is a real time data processing system. These two are completely different use-cases. I know that real-time data would be so much better – but given that Storm needed to be backed by a data store which does aggregation on each of the bolts – will make it likely very slow for processing trillions of words.
Storm would be very good for continuous real time stats on say billions of words.

]]> By: Himanshu Bansal http://gigaom.com/2011/08/04/twitter-to-open-source-hadoop-like-tool/#comment-644838 Thu, 04 Aug 2011 23:50:08 +0000 http://gigaom.com/?p=388794#comment-644838 This seems like an interesting tool.

]]>