7 days left to influence the Highload ++ Siberia program
First about the main thing, and then about everything in order. After 7 days, that is , April 30, the reception of reports on Highload ++ Siberia is closed .
We considered that one HighLoad ++ per year is not enough. Pranked their brains and decided at the same time to expand the geography. Novosibirsk is a large IT-hub in Russia, a number of large IT-companies are based there, and the Siberian Valley, Academpark, is developing. Therefore, we all go to Novosibirsk on June 25 and 26.
Highload ++ Siberia is a fork of our coolest conference for developers of highly loaded systems , but it will be a completely independent event with its future traditions and its program. That is, if something was missing for you all the time on Highload ++, then it's time, or rather the last chance , to actively influence this.
Yes, yes, we expect from you completely new intriguing details about your experience with heavy loads. You can submit a completely original report on a topic that was not previously consecrated, but related to highload, or offer your own format for a meeting, round table, and more.
As always with Highload ++ , topics are not limited to programming only. Issues of operation , testing , client and even mobile development , management and much more will also not be left without attention. But we consider all this exclusively in applications in large projects and under heavy loads .
If your project has millions of users, hundreds of servers, terabytes of data and all this 24/7, then any brick of this system may turn out to be a topic for report - take a closer look!
If you have just a little less, then feel free to apply . In any case, after submitting the abstracts, work on the report is just beginning.
Basic information about how and what to write theses , what stages await you on the way to speaking and what are the conditions for participation are collected on the page for speakers and in our wiki .
Plus, we opened a speaker school in the format of the Telegram channel . There we will publish materials on how to become a cool speaker and how to prepare useful and memorable speeches.
Doubts “I didn’t speak to such large audiences - I'm afraid!” push away. Our head coach-speaker, Poborchy, conducted a study “ Is it easy to speak at a conference for the first time ”. And it turned out that the devil is not so terribleas they paint him. If you missed it, then we recommend you to pay attention, Roman cited statistics and analyzed the most common mistakes of beginning speakers . If you carefully follow the recommendations, then everything will be fine.
And by the way, the argument "Novosibirsk is so far away" is also not accepted. We knocked out discounts from a couple of hotels and help to buy accommodation cheaper than market prices, and we agreed with Aeroflot on pleasant bonuses for conference participants.
And for the speakers, we try to create the most comfortable conditions and take on most of the costs, and this is another incentive to submit a report.
We repeat, there are 10 days left until the end of the collection of applications, therefore, of course, the program is not yet ready. But it’s understandable, it’s interesting for you, and we will nevertheless talk about some interesting applications .
Pavel Dimitryuk from Beeline wants to share the experience of using In-Memory Data Grid technology in streaming data processing. He promises to reveal what additional features appear if the stream data window is stored in in-memory. And point out the problems that were discovered during operation.
Software Transactional Memory is an approach for building competitively changing data models. Alexander Granin from Restaumatic suggests considering the use of his cpp_stm_free library for C ++, which implements this concept in a purely functional style .
Using several real-life tasks as an example, we learn about ways to optimize both the transactional data model and the library itself. We will also talk about the problems and features of functional programming in systems with high performance requirements.
Alexander Tokarev from DataArt suggests discussing the need for row level security in principle and telling where and how best to organize it for a highly loaded project. Let's look at an example of a practical case for choosing a row level security implementation method and analyze three RLS implementation technologies in Oracle DBMS .
The theory of algorithms and data structures, unfortunately, is losing popularity among developers. The need to write your own binary search implementation is causing fierce debate. But Roman Vishnevsky (Tinkoff Development Center), as a finalist of the world programming championships, believes that such knowledge is needed, and expects to prove it using a life example when the use of theoretical, even olympiad, approaches helped effective development .
It is easy to see that a large number of modern web projects are switching to microservice architecture . Konstantin Evteev in his report promises to share Avito's experience in implementing dependent data changes in several services without two-phase commits and synchronous dependencies, so that as a result, data integrity is not violated.
Alexander Makarov (CFT Group of Companies) intends to demonstrate a method for identifying bottlenecks in the server side of the software using the Oracle database as an example. The use of this method on a regular basis allowed a 10-fold reduction in the number of performance incidents on a combat database with a CFT Retail Bank.
Stop what is it at all
We considered that one HighLoad ++ per year is not enough. Pranked their brains and decided at the same time to expand the geography. Novosibirsk is a large IT-hub in Russia, a number of large IT-companies are based there, and the Siberian Valley, Academpark, is developing. Therefore, we all go to Novosibirsk on June 25 and 26.
Highload ++ Siberia is a fork of our coolest conference for developers of highly loaded systems , but it will be a completely independent event with its future traditions and its program. That is, if something was missing for you all the time on Highload ++, then it's time, or rather the last chance , to actively influence this.
Apply
Yes, yes, we expect from you completely new intriguing details about your experience with heavy loads. You can submit a completely original report on a topic that was not previously consecrated, but related to highload, or offer your own format for a meeting, round table, and more.
As always with Highload ++ , topics are not limited to programming only. Issues of operation , testing , client and even mobile development , management and much more will also not be left without attention. But we consider all this exclusively in applications in large projects and under heavy loads .
If your project has millions of users, hundreds of servers, terabytes of data and all this 24/7, then any brick of this system may turn out to be a topic for report - take a closer look!
If you have just a little less, then feel free to apply . In any case, after submitting the abstracts, work on the report is just beginning.
Speaker School
Basic information about how and what to write theses , what stages await you on the way to speaking and what are the conditions for participation are collected on the page for speakers and in our wiki .
Plus, we opened a speaker school in the format of the Telegram channel . There we will publish materials on how to become a cool speaker and how to prepare useful and memorable speeches.
Doubts “I didn’t speak to such large audiences - I'm afraid!” push away. Our head coach-speaker, Poborchy, conducted a study “ Is it easy to speak at a conference for the first time ”. And it turned out that the devil is not so terribleas they paint him. If you missed it, then we recommend you to pay attention, Roman cited statistics and analyzed the most common mistakes of beginning speakers . If you carefully follow the recommendations, then everything will be fine.
And by the way, the argument "Novosibirsk is so far away" is also not accepted. We knocked out discounts from a couple of hotels and help to buy accommodation cheaper than market prices, and we agreed with Aeroflot on pleasant bonuses for conference participants.
And for the speakers, we try to create the most comfortable conditions and take on most of the costs, and this is another incentive to submit a report.
Program
We repeat, there are 10 days left until the end of the collection of applications, therefore, of course, the program is not yet ready. But it’s understandable, it’s interesting for you, and we will nevertheless talk about some interesting applications .
Processing and storing streaming data in Apache Ignite
Pavel Dimitryuk from Beeline wants to share the experience of using In-Memory Data Grid technology in streaming data processing. He promises to reveal what additional features appear if the stream data window is stored in in-memory. And point out the problems that were discovered during operation.
Combinatorial Software Transactional Memory
Software Transactional Memory is an approach for building competitively changing data models. Alexander Granin from Restaumatic suggests considering the use of his cpp_stm_free library for C ++, which implements this concept in a purely functional style .
Using several real-life tasks as an example, we learn about ways to optimize both the transactional data model and the library itself. We will also talk about the problems and features of functional programming in systems with high performance requirements.
Place of row level security in a high-load project
Alexander Tokarev from DataArt suggests discussing the need for row level security in principle and telling where and how best to organize it for a highly loaded project. Let's look at an example of a practical case for choosing a row level security implementation method and analyze three RLS implementation technologies in Oracle DBMS .
Olympiad programming in real projects
The theory of algorithms and data structures, unfortunately, is losing popularity among developers. The need to write your own binary search implementation is causing fierce debate. But Roman Vishnevsky (Tinkoff Development Center), as a finalist of the world programming championships, believes that such knowledge is needed, and expects to prove it using a life example when the use of theoretical, even olympiad, approaches helped effective development .PG Saga
It is easy to see that a large number of modern web projects are switching to microservice architecture . Konstantin Evteev in his report promises to share Avito's experience in implementing dependent data changes in several services without two-phase commits and synchronous dependencies, so that as a result, data integrity is not violated.
Proactive Oracle Database Performance Optimization
Alexander Makarov (CFT Group of Companies) intends to demonstrate a method for identifying bottlenecks in the server side of the software using the Oracle database as an example. The use of this method on a regular basis allowed a 10-fold reduction in the number of performance incidents on a combat database with a CFT Retail Bank.
In our opinion, the prospect of spending a summer weekend with a bunch of like-minded people is even worth it in itself to join.
You can apply here .
April 30 is the last day of filing applications for Highload ++ Siberia ,
May 20 is the deadline for deciding whether to include the report in the program.
You can book a ticket here .
By the way, after April 30, the cost of participation will increase .
