About Aswanth11

Aswanth11 · ‎03-12-2018

Is it possible to write into a location directly instead of hdfs path ?

Aswanth11 · ‎07-25-2016

Is there way to configure and send mail whenever hadoop environment is down, 1. Hive is down 2. One of data node is down or its space reached its threshold

Aswanth11 · ‎04-25-2016

@Benjamin Leonhardi All i need to do is on Hive external tables directly. 1.My above DDL statement was not working when i try to create parquet external table with snappy compression. 2. Is there a way to alter compression from snappy to ZIP in an existing hive external table.

Aswanth11 · ‎04-22-2016

Throwing this error while creating Hive parquet table with snappy compression in hive beeline mode. Error: Error while compiling statement: FAILED: ParseException line 19:15 cannot recognize input near 'parquet' '.' 'compress' in table properties list (state=42000,code=40000) CREATE EXTERNAL TABLE testsnappy ( column bigint ) row format delimited fields terminated by ',' STORED as PARQUET LOCATION 'path' TBLPROPERTIES ("parquet.compress"="SNAPPY") " ; Also is there a way to set compression format for already created tables ?

Aswanth11 · ‎03-18-2016

Thanks much. Name node uses only RAM memory?..no space is required as such data node ?

Aswanth11 · ‎03-17-2016

our HDFS cluster size is 16 TB. have 3 data nodes and 1 name node. 1.how to find out the storage size alloted for 3 data nodes ? 2. Does name node uses only RAM memory ? how to find its capacity as well

Aswanth11 · ‎03-03-2016

Yes Tim. That works. Here we need to join target table (table2). If CREATE SEQUENCE function is there in impala, it would be of more useful instead of doing these tricky things. Thanks for your reply.

Aswanth11 · ‎03-02-2016

Thanks Tim for answering it. I tried using UDFrowsequence java code and it always start from 1. http://svn.apache.org/repos/asf/hive/trunk/contrib/src/java/org/apache/hadoop/hive/contrib/udf/UDFRowSequence.java Is there a way to start from max value by passing it ?

Aswanth11 · ‎03-01-2016

whenever a new record is inserted, sequence number has to be generated for key columns. first day : 100 record with sequence number from 1 to 100. second day : 10 new records. Sequence number should start from 101 to 110. We can achieve this in ETL tools. Also, In some databases, we have create sequence function and use the next value in select clause. Is it possible to achieve in impala ?

Aswanth11 · ‎02-29-2016

I want to generate sequence number for each new record in Impala. 1. First load - am loading 100 records with sequence number starting from 1 to 100. 2. Second load - i need to load new 10 records which are not present in first load . So my sequence number has to start from 101 to 110 for these new to records. I think there is no sequence function here and Next value cannot be used to hold the max value. Is this possible with any other function or UDFs has to be written ?

Online	Offline
Last Visited	‎03-13-2018 03:39 AM

Member Since	‎02-29-2016 11:42 PM
Last Visited	‎03-13-2018 03:39 AM
Posts	41
Kudos received	20

Cloudera Community

Cloudera Community

Re: Generating File by reading Hive data in Spark

Hadoop environment monitoring

Re: Hive table format and compression

Hive table format and compression

Re: HDFS capacity

HDFS capacity

Re: Sequence number generation in impala

Re: Sequence number generation in impala

Re: Is it possiable Auto Increment columns in Impa...

Sequence number generation in impala