Possible to get counters such as Spill file numbers from mapreduce log?

2014-03-03 Thread Felix .
Hi all, I'm wondering if it is possible to get information such as Spill file numbers, Spill start/ end time for each Mapper from mapreduce's log? Thanks!

Collect, Spill and Merge phases insight

2013-07-17 Thread Felix .
Hi all, I am trying to understand the process of Collect, Spill and Merge in Map, I've referred to a few documentations but still have a few questions. Here is my understanding about the spill phase in map: 1.Collect function add a record into the buffer. 2.If the buffer exceeds a threshold

Collect, Spill and Merge phases insight

2013-07-16 Thread Felix .
Hi all, I am trying to understand the process of Collect, Spill and Merge in Map, I've referred to a few documentations but still have a few questions. Here is my understanding about the spill phase in map: 1.Collect function add a record into the buffer. 2.If the buffer exceeds a threshold

Policies for placing a reducer

2013-07-15 Thread Felix .
Hi all, What is the policy of choosing a node for a reducer in mapreduce (Hadoop v1.2.0)? For example, If a cluster has 5 slaves, each slave can serve 2 maps and 2 reduces , there is a job who occupies 5 mappers and 3 reducers , how does the jobtracker assign reducers to these nodes (choosing

Re: Job level parameters

2013-07-01 Thread Felix .
, and there are detailed description. On Tue, Jul 2, 2013 at 11:14 AM, Felix.徐 ygnhz...@gmail.com wrote: Hi all, Is there a detailed list or document about the job specific parameters of mapreduce ? Thanks!

Re: CombineFileInputFormat ran out of memory while making splits

2012-07-16 Thread Felix .
My hadoop version is 1.0.1 and I didn't specify any parameter. 2012/7/16 Felix.徐 ygnhz...@gmail.com HI all, I have written a MyCombineFileInputFormat extends from CombineFileInputFormat , it can put multi files together into the same inputsplit, it works fine for just a small amount of files

How to run or submit MapReduce Job to Hadoop in my own program?

2011-05-17 Thread Felix .
Hi,all..How can I run a MR job though my own program instead of using console to submit a job to a real Hadoop env? I write code like this, this program works fine but i don't think it ran in my Hadoop env,since nothing was produced in hadoop logs folder. public int run(String[] args) throws