Given the odd white stuff here in Vancouver here in the first week of December, the memory of the unseasonably warm NYC in late October of the recent Strata/Hadoop-World event in the Big Apple.

 

The team at Oreilly has now released the keynotes (free) and the video compilation of the sessions (pay-wall).

 

The theme of this Strata shows the pace of the Big Data/Hadoop market place. I can remember just two years ago when the focus was to educate and justify Hadoop. This year’s theme of Hadoop as the new enterprise data hub reflects the maturity of practitioners and also the technologies themselves.

 

But it is December and we’re all about to be deluged with 2013 retrospectives. So I’ll kick it off here with a review of 2013 through my Hive-color glasses.

  1. January kicked off with Hortonworks HDP 1.2 which brought Hive 0.10 and also HiveServer2. I’m proud to say that Simba shipped the first HiveServer2 ODBC driver for Hortonworks.
  2. February was Strata Santa Clara where Intel announced IDH. IDH includes both Simba’s Hive and HBase ODBC drivers.
  3. The April Hive User Group meeting saw the beginning of a number of Stinger deliverable making it out into the wild. HIVE-3746 is recognized as a significant problem for production uses for HiveServer2.
  4. It was a busy summer with a lot of activities in Stinger and elsewhere. The three meetups (June @SF, June @Hadoop Summit, and July) provided a plethora of real-world scenarios and emerging technologies.
  5. Cloudera and Microsoft joined the ranks of leading Hadoop distro suppliers who ships Simba’s industrial-strength Hive ODBC driver. Cloudera CDH4.2 also included Simba’s Impala ODBC driver.
  6. The summer focused on enterprise-grade features—notably authentication and authorization. Simba supports MapR with adding LDAP support into the ODBC driver.
  7. The pre-announcements of Tableau for OS X was probably the hottest tip coming out of the Tableau Customer Conference. The fall ushered in longer night and a 64-bit Tableau 8.1. Simba’s Hive ODBC drivers are there to provide unparalleled access to even larger dataset.