InfluxDB is time series database that can store huge amount of data for different metrics like CPU, RAM, Storage usage. If you have Grafana in place, one of the databases that would work with it is InfluxDB.

If you setup InfluxDB you can type influx and hit enter to see InfluxDB console (like MySQL):

infladmin@my_host:~$ influx
 Connected to http://localhost:8086 version 1.2.2
 InfluxDB shell version: 1.2.2

Databases in InfluxDB are called well Database :) and tables are called Measurement. To see list of all databases and make a database active do as below:

> SHOW DATABASES
 name: databases
 name
 ----
 _internal
 my_db
 monitoring
 some_other_db

 > USE my_db
 Using database my_db

You are now on my_db database. To display list of measurements (table) and query it:

> SHOW MEASUREMENTS
 name: measurements
 name
 ----
 cpu_stat
 mem_stat

 > SELECT * FROM cpu_stat LIMIT 2;
 name: cpu_stat
 time                my_tag      metric         value
 ----                ----------- ------         -----
 1503999476000000000 CPU_CORE3   CORES            0
 1503999476000000000             total            16

Yes, it is almost similar to MySQL. The main point of such a database is to use monitoring tool in place and push metrics into influxDB, then make graphs by using Grafana.

#influxdb #grafana #measurement #time_series_db

Shift time in python using datetime & timedelta:

import datetime
 from datetime import timedelta

 date1 = datetime.datetime.utcnow()
 # output of date1: datetime.datetime(2017, 11, 8, 10, 24, 25, 19492)
 date2 = date1 + timedelta(days=1)
 # output of date2: datetime.datetime(2017, 11, 9, 10, 24, 25, 19492)

To get your new date in EPOCH:

at_epoch = (date2 - datetime.datetime(1970, 1, 1)).total_seconds()
 # output of subtract: 1510223065.019492

To turn EPOCH to datetime:

import time

 from_epoch = time.localtime(1510223065.019492)
 # output: time.struct_time(tm_year=2017, tm_mon=11, tm_mday=9, tm_hour=13, tm_min=54, tm_sec=25, tm_wday=3, tm_yday=313, tm_isdst=0)

 formatted_date = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(1510223065.019492))
 # output: 2017-11-09 13:54:25

#python #epoch #datetime #time #timedelta #strftime

O(1) or O(n) access time?

O(1) means that it takes a constant time, like 14 nanoseconds, or three minutes no matter the amount of data in the set.

O(n) means it takes an amount of time linear with the size of the set, so a set twice the size will take twice the time. You probably don't want to put a million objects into one of these.

#time #complexity #time_complexity

If you have multiple statements and you want to time the whole statements in a bash script you can use time command, but you don't need to use inside of the script. If you have an script called my_long_tasks.sh you need to just:

$ time my_long_tasks.sh
 real    0m20.894s
 user    0m3.664s
 sys     0m0.452s

That's it. In case you want to just time part of your statements in your bash script then you need to:

start=`date +%s`
 stuff
 end=`date +%s`

 runtime=$((end-start))

#linux #bash #time