Factors that limit speed Presto? - hive

I have just installed Presto today on our server at work (version 0.57) and when doing a select count(*) from log; it takes more than 17 minutes for a table with only 640 million records (~64GB).
Now I am under the impression that this is way too slow for presto, but I am not sure.
Some info:
Hive and Presto have both been installed with default configurations from their documentation.
Hive table is an external table with about 24 columns most of them String and 3 of them are Array and the file is stored as Textfile (Hive complains about RCFile with my file for some reason).
The table will be mostly used for grouping and count operations.
Do you have any tips for increasing performance or what the targetted query time should be for a simple count(*) of a table?
Cheers

You should solve your problem with RCFile. Using RCFile will increase the performance significant (x2 - x4 the developers say conform with my experience). Try to convert it using CREATE TABLE <new rcfile table name> AS SELECT * FROM <old textfile table name>; in Presto. (Be sure to have enough space on disk.)

Related

PutHiveQL NiFi Processor extremely slow - misconfiguration?

I am currently setting up a simple NiFi flow that reads from a RDBMS source and writes to a Hive sink. The flow works as expected until the PuHiveSql processor, which is running extremely slow. It inserts one record every minute approximately.
Currently is setup as a standalone instance running on one node.
The logs showing the insert every 1 minute approx:
(INSERT INTO customer (id, name, address) VALUES (x, x, x))
Any ideas about why this may be? Improvements to try?
Thanks in advance
Inserting one record at a time into Hive will result extreme slowness.
As your doing regular insert into hive table:
Change your flow:
QueryDatabaseTable
PutHDFS
Then create Hive avro table on top of HDFS directory where you have stored the data.
(or)
QueryDatabaseTable
ConvertAvroToORC //incase if you need to store data in orc format
PutHDFS
Then create Hive orc table on top of HDFS directory where you have stored the data.
Are you poshing one record at time? if so may use the merge record process to create batches before pushing into HiveQL,
It is recommended to batch into 100 records :
See here: https://nifi.apache.org/docs/nifi-docs/components/org.apache.nifi/nifi-hive-nar/1.5.0/org.apache.nifi.processors.hive.PutHiveQL/
Batch Size | 100 | The preferred number of FlowFiles to put to the database in a single transaction
Use the MergeRecord process and set the number of records or/and timeout, it should speed-up considerably

Oracle Bulk Insert Using SQL Developer

I have recently taken data dumps from an Oracle database.
Many of them are large in size(~5GB). I am trying to insert the dumped data into another Oracle database by executing the following SQL in SQL Developer
#C:\path\to\table_dump1.sql;
#C:\path\to\table_dump2.sql;
#C:\path\to\table_dump3.sql;
:
but it is taking a long time like more than a day to complete even a single SQL file.
Is there any better way to get this done faster?
SQL*Loader is my favorite way to bulk load large data volumes into Oracle. Use the direct path insert option for max speed but understand impacts of direct-path loads (for example, all data is inserted past the high water mark, which is fine if you truncate your table). It even has a tolerance for bad rows, so if your data has "some" mistakes it can still work.
SQL*Loader can suspend indexes and build them all at the end, which makes bulk inserting very fast.
Example of a SQL*Loader call:
$SQLDIR/sqlldr /#MyDatabase direct=false silent=feedback \
control=mydata.ctl log=/apps/logs/mydata.log bad=/apps/logs/mydata.bad \
rows=200000
And the mydata.ctl would look something like this:
LOAD DATA
INFILE '/apps/load_files/mytable.dat'
INTO TABLE my_schema.my_able
FIELDS TERMINATED BY "|"
(ORDER_ID,
ORDER_DATE,
PART_NUMBER,
QUANTITY)
Alternatively... if you are just copying the entire contents of one table to another, across databases, you can do this if your DBA sets up a DBlink (a 30 second process), presupposing your DB is set up with the redo space to accomplish this.
truncate table my_schema.my_table;
insert into my_schema.my_table
select * from my_schema.my_table#my_remote_db;
The use of the /* +append */ hint can still make use of direct path insert.

How to get count record in hbase table? which is fastest way to query the record?

I have 100 Million record in HBase table. I have created hive external table.
How to query the record fastest way.
Hive ---> Select count(*) from table.
Running Query more than 8 hours.
Please guide me
I think the better way here would be use Hbase in built RowCounter operation which internally runs a map reduce job to count the number of rows.
Syntax:
hbase org.apache.hadoop.hbase.mapreduce.RowCounter mytable
Hive supports COUNT() query directly-
SELECT COUNT(*) FROM table
But it will get slow as your records increases because hive uses MapReduce jobs. If you want to query really fast, I would recommend you using Apache Phoenix or ORM tool Kundera

Schema Evolution in Parquet Hive table

I have a lot of data in a Parquet based Hive table (Hive version 0.10). I have to add a few new columns to the table. I want the new columns to have data going forward. If the value is NULL for already loaded data, that is fine with me.
If I add the new columns and not update the old Parquet files, it gives an error and it looks strange as I am adding String columns only.
Error getting row data with exception java.lang.UnsupportedOperationException: Cannot inspect java.util.ArrayList
Can you please tell me how to add new fields to Parquet Hive without affecting the already existing data in the table ?
I use Hive version 0.10.
Thanks.
1)
Hive starting with version 0.13 has parquet schema evoultion built in.
https://issues.apache.org/jira/browse/HIVE-6456
https://github.com/Parquet/parquet-mr/pull/297
ps. Notice that out-of-the-box support for schema evolution might take a toll on performance. For example, Spark has a knob to turn parquet schema evolution on and off. After one of the recent Spark releases, it's now off by default because of performance hit (epscially when there are a lot of parquet files). Not sure if Hive 0.13+ has such a setting too.
2)
Also wanted to suggest to try creating views in Hive on top of such parquet tables where you expect often schema changes, and use views everywhere but not tables directly.
For example, if you have two tables - A and B with compatible schemas, but table B has two more columns, you could workaround this by
CREATE VIEW view_1 AS
SELECT col1,col2,col3,null as col4,null as col5 FROM tableA
UNION ALL
SELECT col1,col2,col3,col4,col5 FROM tableB
;
So you don't actually have to recreate any tables like #miljanm has suggested, you can just recreate the view. It'll help with the agility of your projects.
Create a new table with the two new columns. Insert data by issuing:
insert into new_table select old_table.col1, old_table.col2,...,null,null from old_table;
The last two nulls are for the two new columns. That's it.
If you have too many columns, it may be easier for you to write a program that reads the old files and writes the new ones.
Hive 0.10 does not have support for schema evolution in parquet as far as I know. Hive 0.13 does have it, so you may try to upgrade hive.

bcp out or select query with output in a file, which is faster?

I need to unload around 5-6 million rows into a file from a sybase ASE database table. What is the best way to do that: bcping out or select * from... and storing the output to the file?
The table has some indexes on it. The database server is on a different machine than the file needs to be created.
Any ideas how can it be made faster?
The BCP utility is designed for that purpose. It should be faster than any select *, particularly if you use the native mode, and not the character mode.