
In this issue
- Car giant invests hundreds of millions of dollars in
Pivotal-
Akka is not the same
-
Apache Spark will have to make some room
- How to add a new method to
java.lang.Runtime ?
- How to get the source string, knowing only its hash?
… and much more
1. Trends
1.1. Pivotal again raised the dough
As it became known,
Microsoft and
Ford invested about $ 250 million in
Pivotal . Microsoft's motivation is not particularly interesting - IT people invested in IT people. Much more interesting task Ford. A few comments from the auto giant's CIO,
Marcy Klevron , which will join the Pivotal board:
Our decision is part of our mobility company.
We are re-positioning the company into a software world.
Very indicative situation. IT technologies penetrate an increasing number of sectors of the real sector of the economy, and have every chance to take key positions there in the future for several decades. I think in 2100 it will be very difficult to say which department is most important for a typical automotive giant - designers, designers or IT specialists?
“800 horses, 1000km without recharging, Android Microservices Edition sport package, not a bit, not beautiful, bargaining at the hood ...”
1.2. Where is Akka?
Thoughtworks removed Akka from its radar, “archiving” this technology. In fact, this means "Akka has become boring." In my opinion, the decision is correct. Many words have already been said about the model of actors and reactive programming, and many letters have been written. The pros and cons are clear, the scope too. Technology has become adult and mundane.
')
The guys from the Scala-world, of course, are not very happy:
1.3. Google is catching up with Spark
Meet the new project in the
Apache incubator -
Apache Beam . This is a set of interfaces for creating data processing pipeline. You write a program using these interfaces, and then run it on a specific engine, be it
Apache Spark or
Google Cloud DataFlow .

Source:
shinetechblog.files.wordpress.comThe main sponsor of the project (who would have thought?) -
Google . They explain their motivation
in the following
way : make a high standard, drag many applications onto it, and then slip your cloud platform under these applications. For example, with the help of
such benchmarks , hahaha. In fact, a big bet is made on the open-source in general, and Apache in particular, as the most influential OSS-community.
Very timely traffic, which can give Google a good chance to ride the growing trend of
bigdato-processing .
1.4. What is your enterprise?
Plumbr collected statistics on the use of JEE containers. The results are quite expected, most of all are
Tomcat and
JBoss .
2. Read
2.1. Azul poured Runtime.onSpinWait () into OpenJDK
Link :
https://www.azul.com/jep-285-small-perfectly-formed/A rare case when a company with a name other than
Oracle , independently promoted and sponsored a feature that also changes the base class
java.lang.Runtime .
The process was not easy. Especially difficult was the decision to name the corresponding method. In the course of the discussion, the nerves of people have already begun to
fail . From the outside, this may seem like a bureaucratic cost, but in practice the correct naming of the classes and methods of the framework or platform is hardly inferior in importance to implementation. As the boat name, so it will float.
In the end, everything was resolved safely. Now spinning in JDK9 will be more fun.
2.2. An interesting discussion of Java Mission Control
Link :
https://groups.google.com/forum/#!topic/mechanical-sympathy/uJqHLd_i2hEThe thread started with scripting in JMC, but very quickly went into offtop, switching to a discussion of the advantages and disadvantages of JMC, as well as its upcoming changes in JDK9. Interesting.
2.3. An excellent report from a Netflix engineer on the SRE toolbox.
Link :
http://www.brendangregg.com/blog/2016-05-04/srecon2016-perf-checklists-for-sres.htmlSRE is
site reliability engineering . In short - these are devops, performers and architects in one person. Their task is to make the service a) work; b) worked quickly; c) scaled. The task is interesting, but not the easiest. Follow the link for a report on what tools the guys from the SRE department of
Netflix use . Instructive.
3. Wisdom
3.1. In-memory pro
3.2. About moki and too categorical statements
3.3. About "write it all yourself"
3.4. Pro throughput and latency
4. Humor
4.1. Correct Exception Handling
Just add a StackOverflow link to the message.
4.2. Get string by hash
[: |||||:]Both laughter and sin: the developer
asked to help him with the string hashing algorithm. The main requirement is the ability to get the source string by hash.
The answer was not long in coming:
#include <string> int main() { std::string s = "Hai!"; std::string* ptr = &s; // this is a pointer std::string copy = *ptr; // this retrieves the original string std::cout << copy; // prints "Hai!" }
An occasion to reflect on those who do not like the fundamental questions at the interviews.
4.3. It became known who was the prototype of Imp in the first Doom
[: |||||:]This is
Phillip Heath :

Conclusion: the iron, and perhaps someday and you draw a character from the next shooter.
Releases: Next