Welcome to the Tarantool meetup January 28th

On January 28, 2016, the second Tarantool meetup meeting will be held in the Moscow office of Mail.Ru Group. If someone else does not know: Tarantool is an open source NoSQL In-Memory database management system designed to provide the highest possible performance. At the second meeting, we will consider the main advantages and features of Tarantool, tell about our experience in using this product and plans for the future. First of all, this meeting will be of interest to developers, Unix system administrators, and other specialists who work with databases in one way or another. See the meeting program under the cut.

18:30 - gathering of participants.

19:00 - “Tarantula - how we came to develop our own DBMS” , Denis Anikin, Mail.Ru Group

I will talk about the most common problems of highly loaded projects, about their causes and standard ways to solve them. You will find out why such projects stop working at some point, and how Tarantool can help in this case.

The decision to develop your own general-purpose DBMS (database management system) usually takes a long time to mature. Before rushing into this abyss, we first look for whether it is possible to solve current problems with existing tools, including creating on top of them or integrating into various solutions. And only when a bright image of an ideal database with all the necessary features looms before us, we rush to the worst and begin to create.

This is exactly how Tarantool appeared - one of the fastest databases in the world, which is widely used in Mail.Ru Group and beyond. From the very beginning, Tarantool was developed by a separate team, in open source and aimed at general use, and as a result we got an excellent product - general-purpose DBMS with a unique set of features.

19:40 - “User sessions in Badoo” , Dmitry Novikov, Badoo

Before we started using Tarantool, Badoo user sessions were stored in a proprietary daemon. The demon was good, but there were some problems with it. For example, he stored data only in memory. Anyway, everything that happened with the sessions began to cause certain inconvenience. Therefore, in order for software development to bring joy and pleasure, we decided to radically alter the system that works with sessions.

20:20 - “The practice of using Tarantool in products: the Wallarm experience” , Alexander Golovko, Wallarm

The report will present the practical results of two years of experience using Tarantool in an alienable software solution. We needed to carry out asynchronous analytics according to the data of short-lived objects of HTTP queries. We examined a number of possible solutions to this problem, and as a result we chose Tarantool and the application-server concept. The report will compare the characteristics of all the considered tools and provide arguments and criteria when choosing an architecture. The final part of the presentation is devoted to the problems that developers faced when implementing, operating and migrating to new versions of Tarantool.

When designing our system for protecting web applications, we were faced with the need to correlate data for several already processed HTTP requests. The most effective solution to this problem was to separate the query processing logic between the synchronous process of calculating parameters for analysis and the asynchronous process of analytics itself. In fact, we took out to Tarantool all the post-processing of HTTP requests after their initial analysis on the Nginx side. As a result, it became possible to detect various behavioral attacks on web applications, for example, brute force passwords or codes.

21:05 - end of the meeting.

Registration is already open!