Grand opening ;)

Now I create this blog. Looks fine, isn't it?

Popular posts from this blog

Deploy data to multiple servers with Python

One of easiest way to upload files into multiple servers via ssh is using a script that send data uses scp/sshpass and a list of connection data. For this purpose I like to use a Python language. Firstly, define an array with IP and pasword data in logindata.py file: ip_list = [( '5.1.1.2' , 'password1' ), ( '5.1.1.3' , 'password2 ' ), ... ( '5.1.1.N' , 'passwordN ' )] Using separate file for logina data is preferred because you can use multiple similar files with different data, and/or multiple scripts with uses that data to connect. Okay, now we can create upload script (e.g. multi-scp.py ) itself: from subprocess import call from logindata import ip_list user = 'root' local_path = 'uploads/*' remote_path = '/root/' for (ip, password) in ip_list: print ( "-------------- \n Moving to %s" % ip) call([ 'sshpass -p%s scp -r %s %s@%s:%s' % (password, local_...
Read more

What is GPCL?

GPCL is an object-oriented library for PHP, developed by me and used last ears. Today I write an introduction about GPCL . In few words, with GPCL you can write: include("documents.php"); class MyDoc extends TDocument { function __construct($title) { parent::__construct($title); $this->body->AddHeading(1)->content = "Hello, World!"; } } $doc = new MyDoc("Welcome!"); $doc->PrintAll(); On output you get well-formatted HTML document with "Hello, World" heading and "Welcome" Title.
Read more

One billion records table in Postgresql

I have a simple table in Postgersql database with character key. CREATE TABLE public.zero_hids ( hexblock character(32) COLLATE pg_catalog."default" NOT NULL, hid bigint, CONSTRAINT zero_hids_pkey PRIMARY KEY (hexblock) USING INDEX TABLESPACE alcal ) It is required to insert into this table 1000000000 unique records by loading from CSV file. We can split this file to 5 parts by 200 millions each and insert it from command line: !/bin/bash time psql --port=5433 -d db1 -c "COPY zero_hids FROM '/tmp/table0.csv' delimiter AS ',' HEADER CSV" ... time psql --port=5433 -d db1 -c "COPY zero_hids FROM '/tmp/table4.csv' delimiter AS ',' HEADER CSV" Size of each part is around 9 Gigabytes. Allowed memory limit to Postgresql is 32Gb. Here is results: COPY 200000000 real 55m16.461s user 0m0.040s sys 0m0.008s COPY 200000000 real 89m48.861s user 0m0.048s sys 0m0.016s COPY 200000000 real 202m21.283s user 0m0.04...
Read more