<?xml version="1.0" encoding="utf-8"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title>Disqus - Latest Comments for ndimiduk</title><link>http://disqus.com/by/ndimiduk/</link><description></description><atom:link href="http://disqus.com/ndimiduk/comments.rss" rel="self"></atom:link><language>en</language><lastBuildDate>Tue, 02 Jan 2018 22:10:23 -0000</lastBuildDate><item><title>Re: HBase via Hive, Part 2 - Nick Dimiduk</title><link>http://www.n10k.com/blog/hbase-via-hive-pt2/#comment-3689145593</link><description>&lt;p&gt;Please read about HBase index structure. Data is indexed, and thus most quickly accessed in order of data coordinates -- rowkey, column family, column qualifier, &amp;amp;c. This has been written about extensively. I don't know the details of your query or how Hive is building an HBase request from it, but it sounds like you're experiencing a full table scan, or a partial scan that's significantly long. To fetch results faster, include your query constraint in the rowkey!&lt;/p&gt;&lt;p&gt;Hive over sequence files is a question for the hive user list, not these blog comments. You'll get much more qualified answers to Hive questions from there.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Tue, 02 Jan 2018 22:10:23 -0000</pubDate></item><item><title>Re: HBase via Hive, Part 2 - Nick Dimiduk</title><link>http://www.n10k.com/blog/hbase-via-hive-pt2/#comment-3001064789</link><description>&lt;p&gt;You can define an HBase table as a Hive external table, that's part of what's outlined in this post. For Hive/Phoenix integration, see &lt;a href="http://phoenix.apache.org/hive_storage_handler.html" rel="nofollow noopener" target="_blank" title="http://phoenix.apache.org/hive_storage_handler.html"&gt;http://phoenix.apache.org/h...&lt;/a&gt;&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Mon, 14 Nov 2016 00:45:46 -0000</pubDate></item><item><title>Re: HBase via Hive, Part 2 - Nick Dimiduk</title><link>http://www.n10k.com/blog/hbase-via-hive-pt2/#comment-2652701023</link><description>&lt;p&gt;Good question. Used to be I think not, but maybe that's changed more recently. Best to ask on the hive-user mailing list.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Sun, 01 May 2016 15:59:12 -0000</pubDate></item><item><title>Re: HBase via Hive, Part 2 - Nick Dimiduk</title><link>http://www.n10k.com/blog/hbase-via-hive-pt2/#comment-2261092684</link><description>&lt;p&gt;It would be better if this lived in Hive, not in a blog post from N years ago. I believe Hive accepts documentation JIRAs, so please open an issue on &lt;a href="https://issues.apache.org/jira/browse/HIVE" rel="nofollow noopener" target="_blank" title="https://issues.apache.org/jira/browse/HIVE"&gt;https://issues.apache.org/j...&lt;/a&gt;&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Fri, 18 Sep 2015 14:50:08 -0000</pubDate></item><item><title>Re: HBase via Hive, Part 2 - Nick Dimiduk</title><link>http://www.n10k.com/blog/hbase-via-hive-pt2/#comment-2257522953</link><description>&lt;p&gt;Sure. (1) create the table in HBase, using HBase shell or similar. (2) define the hive table as EXTERNAL so that Hive does not manage the HBase table's lifecycle (no "two-way bond", as you say). I mentioned EXTERNAL tables in the previous post.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Wed, 16 Sep 2015 18:08:55 -0000</pubDate></item><item><title>Re: HBase via Hive, Part 2 - Nick Dimiduk</title><link>http://www.n10k.com/blog/hbase-via-hive-pt2/#comment-1931525338</link><description>&lt;p&gt;This example is from an older version of both Hive and HBase. The concepts are the same but the details have surely changed.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Fri, 27 Mar 2015 12:19:01 -0000</pubDate></item><item><title>Re: BlockCache Showdown - Nick Dimiduk</title><link>http://www.n10k.com/blog/blockcache-showdown/#comment-1543114656</link><description>&lt;p&gt;You're probably right. There's more work to do here.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Thu, 14 Aug 2014 16:42:30 -0000</pubDate></item><item><title>Re: HBase via Hive, Part 2 - Nick Dimiduk</title><link>http://www.n10k.com/blog/hbase-via-hive-pt2/#comment-1543113259</link><description>&lt;p&gt;Thanks Michael. Maybe you can try out an Apache Hive release?&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Thu, 14 Aug 2014 16:41:41 -0000</pubDate></item><item><title>Re: Measuring Measures - Measuring Measures - Deploying Clojure Services with Crane</title><link>http://measuringmeasures.com/blog/2010/10/11/deploying-clojure-services-with-crane.html#comment-85962844</link><description>&lt;p&gt;Amusing. I always thought of this project as Crane [&lt;a href="http://bit.ly/bGruYh" rel="nofollow noopener" target="_blank" title="http://bit.ly/bGruYh"&gt;http://bit.ly/bGruYh&lt;/a&gt;], not Crane [&lt;a href="http://bit.ly/1495u9" rel="nofollow noopener" target="_blank" title="http://bit.ly/1495u9"&gt;http://bit.ly/1495u9&lt;/a&gt;] :)&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Mon, 11 Oct 2010 12:55:52 -0000</pubDate></item><item><title>Re: The world's leading companies | GDS Publishing</title><link>http://www.busmanagement.com/news/the-worlds-leading-companies/#comment-48559015</link><description>&lt;p&gt;Is there an error in the "Software and services" data? The bar appears to not match the values in the table - market value is a smaller value than assets but is rendered as a larger percentage of the bar.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Wed, 05 May 2010 16:49:17 -0000</pubDate></item><item><title>Re: Big+Data+Is+Less+About+Size%2C+And+More+About%26nbsp%3BFreedom</title><link>http://techcrunch.com/2010/03/16/big-data-freedom/#comment-71220259</link><description>&lt;p&gt;Alex: I believe there exists a whole class of company which is just barely coming into existence, one much like FlightCaster, in fact. This company needs new kinds of tools for easily dealing with their "medium to big" data. That is exactly what I'm building.&lt;/p&gt;&lt;p&gt;I think you are making an assumption which is false: I believe the size of data is no longer tied directly to the size of the company. Look at all the facebook apps out there who gain millions of users in their first month. Small company, "big" data.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Wed, 17 Mar 2010 20:29:56 -0000</pubDate></item><item><title>Re: Working with Java Arrays</title><link>http://ianp.org/2010/01/working-with-java-arrays/#comment-334431238</link><description>&lt;p&gt;How does one specify an array as the parameter to a constructor in a gen-class scenario? I'd like to extend a class which has multiple single-argument constructors, some of which are of type Class[]. What would that look like in clojure?&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Sat, 16 Jan 2010 02:13:06 -0000</pubDate></item><item><title>Re: Heroku | Commercial Launch</title><link>http://blog.heroku.com/archives/2009/4/24/commercial_launch/#comment-8732260</link><description>&lt;p&gt;Congratulations on your launch!  I look forward to doing business with you guys :D&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Mon, 27 Apr 2009 11:48:35 -0000</pubDate></item><item><title>Re: Heroku | The Future of Deployment</title><link>http://blog.heroku.com/archives/2009/2/6/future_of_deployment/#comment-6060168</link><description>&lt;p&gt;You have a really compelling argument here: I don't want to be an admin, I want to write my apps and build a community.  I'll have my cake and eat it too, thank you very much!&lt;/p&gt;&lt;p&gt;Now tell me, how are you planning to manage IP?  My Heroku apps, are they mine?&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Fri, 06 Feb 2009 21:16:12 -0000</pubDate></item><item><title>Re: Goodbye, Bank of America.  Hello, ______?</title><link>http://blog.wesabe.com/2007/01/29/goodbye-bank-of-america-hello-______/#comment-16788335</link><description>&lt;p&gt;I'm a little surprised USAA won't take you.  Their policies on potential member eligibility is much more lenient than it once was.  If you haven't actually called them and asked, I would not count them out.  I'm quite happy with USAA and would recommend them over any lending institution I've ever dealt with and/or heard about.  Give them a call.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Nick Dimiduk</dc:creator><pubDate>Mon, 29 Jan 2007 18:22:07 -0000</pubDate></item></channel></rss>