hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Tom White <...@cloudera.com>
Subject Re: Minutes: Hadoop Contributor Meeting 05/06/2010
Date Fri, 07 May 2010 16:36:42 GMT
Here's my (single) slide about the 0.21 release.

Tom

On Thu, May 6, 2010 at 5:38 PM, Arun C Murthy <acmurthy@gmail.com> wrote:
> # Shared goals
>  - Hadoop is HDFS & Map-Reduce in this context of this set of slides
> # Priorities
>  * Yahoo
>    - Correctness
>    - Availability: Not the same as high-availability (6 9s. etc.) i.e. SPOFs
>    - API Compatibility
>    - Scalability
>    - Operability
>    - Performance
>    - Innovation
>  * Cloudera
>    - Test coverage, api coverage
>    - APL Licensed codec (lzo replacement)
>    - Security
>    - Wire compatibility
>    - Cluster-wide resource availability
>    - New apis (FileContext, MR Context Objs.), documentation of their
> advantages
>    - HDFS to better support non-MR use-cases
>    - Cluster metrics hooks
>    - MR modularity (package)
>  * Facebook
>    - Correctness
>    - Availability, High Availability, Failover, Continuous Availability
>    - Scalability
> # Bar for patches/features keeps going higher as the project matures
>  - Build consensus (e.g. Python Enhancement Process, JSR etc.)
>  - Run/test on your own to prove the concept/feature or branch and finish
>  - Early versions of libraries should be started outside of the project
> (github etc.) e.g. input-formats, compression-codecs etc.
>    - github for all the above
>    - Prune contrib
> # Maven for packaging
> # Tom: hadoop-0.21 (Tom - can you please post your slides? Thanks!)
> # Owen: Release Manager (see slides)
> # Agenda for next meeting
>  - Eli: Hadoop Enhancement Process (modelled on PEP?)
>  - Branching strategies: Development Models
>
> Arun
>
>
>
>
>

Mime
  • Unnamed multipart/mixed (inline, None, 0 bytes)
View raw message