Query

SQL Simplified {4}

by Andrew M
The article I read was entitled “Interactive SQL Query Suggest: Making Databases User-Friendly” by Ju Fan. This article talks about the difficulty of SQL and how it is very daunting to many people who are new to it. Many users do not use SQL because it is too hard for them to learn the language and to use the queries correctly. Ju Fan says that some of the current solutions are to use keywords to help in creating SQL queries. While this is efficient enough for new users it does not work well enough for more advanced users. The author proposes using something known as SQLSugg or SQL Suggestions. This takes the idea of using keywords and then advances it by using pre-made templates. In effect, the user types in a keyword and SQLSugg produces templates of pre-made queries which might meet the user’s needs. SQLSugg allows advanced users to use these templates and makes the creation of queries much easier. read more...

Making SQL Queries Better {1}

by Andrew M
The article I read this week was entitled “Generating Test Data for Killing SQL Mutants: A Constraint-based Approach” by Shetal Shah, S. Sudarshan, Suhas Kajbaje, Sandeep Patidar, Bhanu Pratap Gupta and Devang Vira. This article talks about how extremely advanced and complicated SQL queries are being created all the time and how difficult it is to test these queries. While SQL queries are always tested before being used sometimes it is very hard to test these queries enough to make sure they will always give the desired results. The authors propose that a method called Mutant Testing be used. In effect, by applying Mutant Testing to test SQL queries testers will be able to show if the query will give the desired results. In Mutant Testing the algorithm of a specific query is changed slightly and the query is ran to test if the results come back the same. If the results are different, as they should be, then this in affect proves that the query is written correctly. While testing every possible mutation of a single query would be extremely time consuming. The authors propose making mutations that would test common programming errors. This would test more efficiently and would not just be testing every possible circumstance. read more...

‘Drill’, A New Project From Apache {1}

by Leonardo S
The article I chose for this week is titled “New Apache Project ‘Drill’ Aims to Speed Up Hadoop Queries” by Todd R. Weiss. The topic of this article is the new project taken up by the Apache Software Foundation called ‘Drill’. For those that don’t know, Apache is a community of users that develops open-source software, including the popular OpenOffice. The purpose of its new project is to have a tool that will speed up the use of Hadoop data analysis tools. By letting users do quicker queries of large data sets, Drill promises to be a reliable replacement to similar paid programs such as Google’s Dremel. The article continues by saying that the need for Drill was brought about by increasing user requirements, which basically means that people are tired of waiting more than a few seconds for query search results. read more...

Wikipedia Creates Wikidata {3}

by Michael V
Wikipedia is often known as a fountain of knowledge for uses to access, as it allows people to search for the information about topics they wish to understand, as well as edit and create entries for others to use. However, this entire system relies primarily on the manpower of millions of contributors, with minimal aid from the system itself. This is because Wikipedia contains mostly knowledge readable by human beings, but cannot be understood by the machines themselves. Wikipedia’s next project, Wikidata, attempts to remedy this issue with the solution of creating a database that is not only readable by both people and machines, but editable by both as well. This in essence will allow the database itself to answer queries asked by users in an automatic fashion, which is a massive leap from having users dig through the information themselves. Typically, questions like these are answered by other users in today’s Wikipedia site, and this advancement will aim to change that. Wikidata will also be able to automatically create tables and charts on what the users need. read more...

Google’s BigQuery {3}

by Ronny W
BigQuery is a cloud-based service from Google. It is use for analyzing very large sets of data. It was unveiled at the Google I/O developer conference two years ago, and it is now publicly available after a period of limited-availability testing. BigQuery is an online analytical processing system that is able to process larger amount of data in real time. Pricing can vary according to the usage of query and the amount data stored. BigQuery is not for OLTP(online transaction processing) task. read more...