Imhotep - Data Analytics Platform by Indeed

Sample Data

NASA Apache Web Logs

The sample time-series dataset in nasa_19950801.tsv comes from public 1995 NASA Apache web logs. The file contains data for a single day and is in an Imhotep-friendly TSV format.

A Perl script was used to convert the Apache web log into the TSV format, extracting the following fields:

host When possible, the hostname making the request. Uses the IP address if the hostname was unavailable.
logname Unused, always -
time In seconds, since 1970
method HTTP method: GET, HEAD, or POST
url Requested path
response HTTP response code
bytes Number of bytes in the reply

Here is an example line (or document) from the dataset:

piweba3y.prodigy.com - 807301196 GET /shuttle/missions/missions.html 200 8677

The timestamp 807301196 is the conversion of 01/Aug/1995:13:19:56 -0500 using Perl:

use Date::Parse;
$in = "01/Aug/1995:13:19:56 -0500";
$out = str2time($in);
print "$out\n";

Data for two months are available in these compressed files:
nasa_19950630.22-19950728.12.tsv.gz
nasa_19950731.22-19950831.22.tsv.gz

TSV Data Size (raw uncompressed) Imhotep Data Size
256 MB 19 MB

Source: Internet Traffic Archive

Wikipedia Web Logs

The time-series data in wikipedia_e_20140913.11.tsv.gz is one hour of data from 9/13/2014 for Wikipedia articles beginning with the letter E.

Each document corresponds to a Wikipedia article that was served in that hour:

title Title of the article on Wikipedia
categories+ List of categories in which the article is contained
titleWords+ List of words in the title
linksOut+ List of Wikipedia articles linked by the article
numRequests Number of requests for the article in that hour
bytesServed Number of bytes served for the article in that hour

The most popular E entry in that hour was English_alphabet.

title categories+ titleWords+ linksOut+ numRequests bytesServed
English_alphabet All_Wikipedia_articles_needing_clarification All_articles_needing_additional_references All_articles_with_unsourced_statements Articles_containing_Old_English-language_text Articles_needing_additional_references_from_June_2011 Articles_with_hAudio_microformats Articles_with_unsourced_statements_from_January_2011 Articles_with_unsourced_statements_from_July_2010 Articles_with_unsourced_statements_from_March_2014 English_spelling Latin_alphabets Wikipedia_articles_needing_clarification_from_August_2013 English alphabet A Adjective Aircraft Alphabet_song American_English American_braille American_manual_alphabet Ampersand Anglo-Saxon_futhorc Anglo-Saxons Ansuz_(rune) Apostrophe B Body_cavity British_English Byrhtfert ... 960 21124206
TSV Data Size (raw uncompressed) Imhotep Data Size
2450 GB 272 GB

Source: https://dumps.wikimedia.org/other/pagecounts-raw/ for page counts and https://dumps.wikimedia.org/backup-index.html for all other fields

World Cup 2014 Player Data

The dataset in worldcupplayerinfo_20140701.tsv includes information about players in the World Cup 2014. Since this is not typical time-series Imhotep data, all documents are assigned the same timestamp: 2014-07-01 00:00:00

Each document in the dataset includes information about a single player:

Player String Player’s name.
Age Int Player’s age.
Captain Int Value (1 or 0) indicates whether the player is a captain.
Club String The player’s club when not playing for the national team in the World Cup.
Country String The country the player represents in the World Cup.
Group String The player’s national team belongs to this World Cup group.
Jersey Int The player’s jersey number.
Position String The player’s position.
Rank Int The ranking of the country the player represents.
Selections Int The number of World Cup appearances for this player.
TSV Data Size (raw uncompressed) Imhotep Data Size
45 KB 15 KB

Source: Stack Exchange Network / Open Data
The data are distributed under the creative commons Attribution-Share Alike 4.0 International license. The creator of the data is http://opendata.stackexchange.com/users/3061/bryan. In compliance with this license, the data is hereby attributed to the users and owners of StackOverflow, but not in such a way as to suggest that they endorse Indeed or Indeed’s use of the data.