sort split merge -u unique


 
Thread Tools Search this Thread
Top Forums Shell Programming and Scripting sort split merge -u unique
# 1  
Old 08-31-2011
sort split merge -u unique

Hi, this is about sorting a very large file (like 10 gb) to keep lines with unique entries across SOME of the columns.

The line originally looked like this:

Code:
sort -u -k2,2 -k3,3n -k4,4n -k5,5n -k6,6n file_unsorted > file_sorted

please note the -u flag.

The problem is that this single command is taking more than 12 hours with 10 g of memory and I am looking for a way to speed things up.

I have heard that splitting a large file into subfiles, then sorting each subfile and then merging back together with the sort command can work, but I am imaging this will not work because I'm using the -u flag to keep only unique rows.

maybe split and then sort first WITHOUT the -u flag and then sort -m -u -k2,2 -k3,3n -k4,4n -k5,5n -k6,6n subfiles ?

I'm thinking of ways to test that, but please let me know if you have any ideas for me!

Thank you kindly.

Jonathan

Last edited by radoulov; 08-31-2011 at 06:01 PM.. Reason: Please use code tags for code and data samples, thank you
# 2  
Old 08-31-2011
You need a database, not shell scripting, to manipulate such amount of data ...
# 3  
Old 08-31-2011
Thanks for the tip; we can consider using a database in the future.

Do any readers have an idea how to accomplish this task within the unix / bash framework I am accustomed to?

Thanks.
# 4  
Old 09-01-2011
Hi.

Most file sorting programs will follow the design noted in External sorting - Wikipedia, the free encyclopedia -- i.e. they will create smaller temporary sorted files, then do as many merges as necessary to accomplish a final sort. If you think about it, this is very reasonable: you sort a memory-load of data (as an array, using one of any number of algorithms), write it out to a temporary file until all input files are read, then you read chunks from each sorted temporary to produce a final file. If there too many temporary files, you might need to do an intermediate merge, etc.

I just glanced though a few thousand lines of GNU sort ( part of coreutils, coreutils-8.5/src/sort.c ) and it seems to do the sorting into temporary files.

I think the most you could do to help a system sort is to provide one or more directories for the temporary files on devices different from the input file:
Code:
       -T, --temporary-directory=DIR
              use DIR for temporaries, not $TMPDIR or /tmp; multiple options
              specify multiple directories

-- excerpt from Linux man sort

You could experiment with your own splitting by splitting a sample of your data, using utility split, and then sorting those files. I'd guess that would be slower than sort's internal splitting, but it might be worth a try, at least it would be educational.

You could also use a trace facility to look at the sort while it is going on. That would probably show scratch files being opened and child processes being created to help with the overall process: strace - trace system calls and signals, for example.

Best wishes ... cheers, drl
# 5  
Old 09-01-2011
Make sure you are sorting in the C locale. The other locales can be 10 time slower.
Login or Register to Ask a Question

Previous Thread | Next Thread

9 More Discussions You Might Find Interesting

1. UNIX for Beginners Questions & Answers

Merge 4 bim files by keeping only the overlapping variants (unique rs values )

Dear community, I am facing a problem and I kindly ask your help: I have 4 different data sets consisted from 3 different types of array. On each file, column 1 is chromosome position, column 2 is SNP id etc... Lets say I have the following (bim) datasets: x2014: 1 rs3094315... (4 Replies)
Discussion started by: fondan
4 Replies

2. Shell Programming and Scripting

Sort unique

Hi, I have an input file that I have sorted in a previous stage by $1 and $4. I now need something that will take the first record from each group of data based on the key being $1 Input file 1000AAA|"ZZZ"|"Date"|"1"|"Y"|"ABC"|""|AA 1000AAA|"ZZZ"|"Date"|"2"|"Y"|"ABC"|""|AA... (2 Replies)
Discussion started by: Ads89
2 Replies

3. Shell Programming and Scripting

How to merge two files with unique values matching.?

I have one script as below: #!/bin/ksh Outputfile1="/home/OutputFile1.xls" Outputfile2="/home/OutputFile2.xls" InputFile1="/home/InputFile1.sql" InputFile2="/home/InputFile2.sql" echo "Select hobby, class, subject, sports, rollNumber from Student_Table" >> InputFile1 echo "Select rollNumber... (3 Replies)
Discussion started by: Sharma331
3 Replies

4. UNIX for Dummies Questions & Answers

Print unique lines without sort or unique

I would like to print unique lines without sort or unique. Unfortunately the server I am working on does not have sort or unique. I have not been able to contact the administrator of the server to ask him to add it for several weeks. (7 Replies)
Discussion started by: cokedude
7 Replies

5. UNIX for Dummies Questions & Answers

How to count specific columns and merge with unique ones?

Hi. I am not sure the title gives an optimal description of what I want to do. I have several text files that contain data in many columns. All the files are organized the same way, but the data in the columns might differ. I want to count the number of times data occur in specific columns,... (0 Replies)
Discussion started by: JamesT
0 Replies

6. Shell Programming and Scripting

How to merge columns into lines, using unique keys?

I would really appreciate a sulution for this : invoice# client# 5929 231 4358 231 2185 231 6234 231 1166 464 1264 464 3432 464 1720 464 9747 464 1133 791 4930 791 5496 791 6291 791 8681 989 3023 989 (2 Replies)
Discussion started by: hemo21
2 Replies

7. Shell Programming and Scripting

Awk sort and unique

Input file --------- 12:name1:|host1|host1|host2|host1 13:name2:|host1|host1|host2|host3 14:name3: ...... Required output --------------- 12:name1:host1(2)|host1(1) 13:name2:host1(2)|host2(1)|host3(1) 14:name3: where (x) - Count how many times field appears in last column ... (3 Replies)
Discussion started by: greycells
3 Replies

8. UNIX for Dummies Questions & Answers

split a file with unique sets

This may sound like a trivial problem, but I still need some help: I have a file with ids and I want to split it 'n' ways (could be any number) into files: 1 1 1 2 2 3 3 4 5 5 Let's assume 'n' is 3, and we cannot have the same id in two different partitions. So the partitions may... (8 Replies)
Discussion started by: ChicagoBlues
8 Replies

9. Shell Programming and Scripting

Sort and Unique in Perl

Hi, May I know, if a pipe separated File is large, what is the best method to calculate the unique row count of 3rd column and get a list of unique value of the 3rdcolum? Thanks in advance! (20 Replies)
Discussion started by: deepakwins
20 Replies
Login or Register to Ask a Question