Whats new in Spark this week #3

Its been a busy week in the world of Spark with some interesting announcements. First off there is another AMP camp being hosted at Berkeley http://ampcamp.berkeley.edu/amp-camp-three-berkeley-2013 from August 29th to 30th . Which, fair warning, is in the middle of burning man for those of you who participate in that. More importantly there is a new Spark stable release (0.7.3) and you can read the release notes at http://spark-project.org/spark-release-0-7-3/. This release is mostly bug fixes, however heavy users of the spark shell may wish to usenew shell environment variable ADD_JARS to add JARS to the spark shell workers.

In this weeks checkins

• https://github.com/karenfeng improved the task UI in https://github.com/mesos/spark/commit/872c97ad829ba20e866c4e45054e7d2d05b02042 , https://github.com/mesos/spark/commit/2eea974795dfa2bb79e66496454f36cb499065b0 , ...
• https://github.com/JoshRosen fixed a bug which caused the Java sampleStdev to compute the regular stdev instead.
• https://github.com/viirya fixed a bug with Spark in Yarn standalone mode copying JARs causing task failure https://github.com/mesos/spark/commit/a613628c5078cf41feb973d0ee8a06eb69615bcf
• https://github.com/shivaram added support for providing an weight vector to the logistic regression functions https://github.com/mesos/spark/commit/84fa20c2a135f54745ddde9abb4f5e60af8856d1
• https://github.com/ScrapCodes upgraded Aka from 2.0.3 to 2.0.5 https://github.com/mesos/spark/pull/708