Year comes to an end, ahead long vacation. For many vacation is an excellent opportunity to sit and look around what at us new and interesting occurs in the professional dzhavovsky world now.
In April in Moscow we carried out in Moscow big Java-conference — JPoint 2015. Conference brought together more than one thousand developers on a site, some more hundreds — watched conference online. We experimented also with opening (Dmitry Galkin's lecture about the modern art and programming really shocked many) and with new formats (round tables and expert discussions). But reports were and remain a key subject of conference.
Videos of all reports of conference lie on Youtube. We, as always, collected statistics from responses of participants and counted a rating of reports. In this post — the traditional overview of the best reports of conference. I will make the short overview of ten best reports of conference you knew about them a little more and looked at what is interesting to you.
Sergey Kuksenko, Oracle — Iron counters on guards of performance
Average score: 4.28
This report won a special prize of jury in the nomination "azzky hardcore". The general idea of the report is as follows: provide that you already naoptimizirovat everything that is possible in the application — looked at a network, OS, JVM, etc. and understood that everything rested against the processor. After that we poprofilirovat, began to work quicker, but all the same the processor is loaded for 100%. What to do?
It appears, in the processor there are different counters of events. This Hardware Performance Counters mechanism is called. The architecture of modern processors is very difficult, in them there can be a lot of miscellaneous. Focus is that we can turn on some counters in the processor which will consider the number of the taken place events. That is, some iron profiler in the processor.
What events this processor profiler is able to consider? Yes practically any. In modern intelovsky processors, according to Sergey, their about one thousand. If you want to understand what events should be looked first of all, where in general to dig and what with it there are difficulties — surely look at this report.
This year Sergey will arrive to us to Moscow again — the truth not from St. Petersburg any more, and from California. It will decide on a subject in January. Most likely it will be something about optimization of performance again.
Baruch Sadogursky, JFrog — How to write asynchronous multithreaded HTTP-applications
Average score: 4.29
In this report Baruch stated some experience which they with guys from JFrog saved up in the course of writing of the systems. If to speak technical terms — they write servers, one of the main tasks of which — quickly to distribute binarnik. That is, multithreaded HTTP-applications, on Java, asynchronous.
The report broke into two parts. In the first Baruch told about how they at themselves selected libraries for asynchronous work with HTTP. What principles make sense to guided at the choice? How to compare performance of such libraries and whether it makes sense? What library to select, eventually?! In the second part of the report Baruch told that not so with HTTP 1.1 and than the future HTTP/2 is good and as it will help all of us.
In April Baruch will arrive to us on JPoint again. And too from California. But the main thing not it, but the fact that Baruch decides on a subject of the report now and takes the corresponding vote. Who gathers for JPoint — click this link and help us with the choice of reports.
Nikita Salnikov-Tarnovsky, Plumbr — Where my memory, the dude?!
Average score: 4.38
Not really difficult and well structured report on how to behave if yours prodakshen will read to fall out with OutOfMemoryError. In the first part of the report Nikita tells very much about what regions of memory are in Java why they are necessary and as garbage collectors work with them. Further the speech went about memory leak symptoms, and also about what is and that is not leak. In the last part of the report Nikita does demos on MAT (Memory Analyzer Tool) and shows how this tool helps with real life. Nothing especially difficult, but is very effective in everyday life.
Nikita is not sure yet whether he will be able to arrive on JPoint this time — for the end of April it already planned issue. But we still, as they say, will fight. Blackmail, threats, tortures — and here it already speaks at conference! As you understood, this issue is almost resolved.
Vladimir Sitnikov, to NetCracker — we Are expressed regularly
Average score: 4.40
The report on regular expressions and on several courses which can well improve performance of your program if you understood that it brakes at the time of use regekspa. And about a heap of a rake which is spread out in this field.
Parsing of texts regulyarka, parsing XML, the most widespread errors, the analysis of that as where can brake. Did not do also without performance tests of different regesp on fighting examples.
On JPoint 2016 Vladimir will tell about what typical problems arise with a performance of different opensorsny solutions which very much like to use in modern Java-projects — Spring, Postgres JDBC Driver, Wildfly, etc. From the practical point of view, I believe, it will be very interesting.
Anton Keks, Codeborne — Architecture of Internet bank without Enterprise
Average score: 4.43
Anton Keks, first of all, is known in Russia for the well-known criticism of "bloody Enterpise" (part 1, part 2). To be exact, criticism of the fact that most of developers drag to themselves in new projects different heavy technologies, practically without thinking that practically it is always possible to manage something significantly more lightweight and simple in use.
In this report Anton told about what lightweight technologies they at themselves in Codeborne use when developing Internet banks and why these technologies are pleasant to them more, than the commonly accepted stacks in the industry based on Java EE, Spring and other remarkable solutions.
Than we will be pleased by Anton on JPoint 2016 in April — it is still unknown. But we work on it. In January it will be clear.
Well and minute of advertizing at the end of a post. The ticket for JPoint 2016 — a good gift for new year to you, your friends and colleagues. We put a set of efforts and therefore I guarantee to you that you will not be disappointed in each our conference.
Found under the Fir-tree the ticket on #jpoint. Now in doubts roofing felts the letter reached Ded, whether the girl overheard? On Ded there is not enough hope. :-(— Ilya Kuchmin (@ikuchmin) on December 28, 2015
Preferential registration for JPoint comes to an end already tomorrow. Therefore do not delay purchase.
This article is a translation of the original post at habrahabr.ru/post/274307/
If you have any questions regarding the material covered in the article above, please, contact the original author of the post.
If you have any complaints about this article or you want this article to be deleted, please, drop an email here: firstname.lastname@example.org.
We believe that the knowledge, which is available at the most popular Russian IT blog habrahabr.ru, should be accessed by everyone, even though it is poorly translated.
Shared knowledge makes the world better.