Applying by sending us your CV to: hr+engineer+web@youthdev.net

Working at Vietnam office: Lv5 - No 06 La Bonita Building, 06 - 08 D2 (Nguyen Gia Tri), Binh Thanh, Ho Chi Minh

Top 3 Reasons To Join Us

We are looking for a Python Engineer responsible for managing back-end services and the interchange of data between the server and the users. Your primary focus will be the development of all server-side logic, definition and maintenance of the central database, and ensuring high performance and responsiveness to requests from the client sides. Therefore, a basic understanding of API technologies is necessary as well. Currently, our back-end system need some knowledge about using FFMPEG for video processing and understanding about running long processing processes and keep it stable, even few minutes downtime or interruption may not be acceptable.

Your Skills and Experience

The array mentioned here is automatically-reallocated array (vector in C++, ArrayList in Java or List in C#).

The benchmark graph between vector & list I showed in this post belongs to “Baptiste Wicht” in http://baptiste-wicht.com/posts/2012/12/cpp-benchmark-vector-list-deque.html (Thanks for saving my time for the benchmarking stuff).

From theory

People say Linked-list has much better performance than array when it comes to random-insertion & random-deletion. That’s what we’ve learn in theory, too.

And I’ve seen many people around the internet suggest to use linked-list when we need to do a lots of random-insertion & random-deletion.

To practice

“How could it be?”

“Impossible!”

Note: The “list” above is implemented as doubly linked-list in C++. And the “vector” is implemented as automatically-reallocated array in C++.

It’s not that what people say is not true. It was once true in theory, and in those old days – when the CPU speed was too slow. But now, the context has changed, and the architecture of computers have changed. Therefore, it might not be true anymore in some situations.

The CPU speed on modern computers are too fast, much faster than RAM-access speed. It would take about several hundreds of CPU cycles to read from RAM (Way too slow!!!). If that’s the case, then most of the time, what the CPU do is just to sit there, drink tea, and wait for its best friend – RAM.

And then, CPU caches come into play

The RAM-access speed is too slow that if CPU processes data directly on RAM, 99% of CPU speed will become a waste, and we can just make use of 1% of its power. Therefore, they invented CPU cache to solve this problem. The cache memory is very fast, and is integrated directly on CPU. Its access speed is near the CPU speed. So now, instead of accessing data directly on RAM, CPU will access data on RAM indirectly through L1 cache (There are usually three levels of caches, and L1 cache is the fastest among them).

However, the CPU cache does not solve the problem completely. Because memory cache size is much smaller than RAM. We still need RAM as our main memory. And the CPU caches will just hold small pieces of data that is most likely to be needed by the CPU in the near future. Sometimes, the data that the CPU need to access next is not already in L1 cache (not in L2 or L3 cache, too), and it must be fetched from RAM, then the CPU will have to wait for several hundreds cycles for the data to become available (This is what we call cache miss).

Therefore, to reduce cache miss, when the CPU wants to access data at address x in RAM, it will not only fetch the data at address x, but also neighborhood of address x. Because we assume “if a particular memory location is referenced at a particular time, then it is likely that nearby memory locations will be referenced in the near future.”[3] (This is what we call locality of reference). So, if the data to be processed by CPU is placed right next to each others, we can make use of locality of reference and reduce cache miss (Which might cause huge performance overhead if occurs often).

The problem of linked-list

Unlike array which is a cache-friendly data structure because its elements are placed right next to each other, elements of linked-list can be placed anywhere in the memory. So when iterating through linked-list, it will cause a lots of cache miss (since we can’t make use of locality of reference), and introduce a lots of performance overheads.

Performance of array vs linked list

When we perform random-insertion or random-deletion on array, the subsequent elements needs to be moved.

However, when it come to a list of small elements (list of POD type, or list of pointers), the cost of moving elements around is so cheap, and much cheaper than the cost of cache misses. Therefore, most of the time, when working with a list of small data types (whatever type of operations it is: from iterating, random-insertion, random deletion to number crunching), performance of array will be much better than linked-list.

But when we need to work with a list of large elements ( > 32 bytes), the cost of moving elements around grows up to be higher than the cost of linked-list’s “usually-happened” cache misses. Then the linked list will have better performance than array.

Conclusion

We should prefer array over linked-list when working with a list of small elements (a list of POD type, a list of pointers in C++ or a list of references in Java/C#).

And nearly most of the time, we will use array because:

• In C++, we don’t often store a list of large data type, but a list of pointers that point to large data type. (If we do need to store a list of large-sized class/struct and need to do a lots of random-insertion and random-deletion, then linked-list would be a better choice)
• In Java, we’re forced to use a list of references that point to object in the heap, instead of a list of objects, because reference type are placed on the heap by default and we can just make a reference to it. (Since reference is just a managed pointer, its size is no more than 8 bytes)
• In C#, we often use class instead of struct, so we also often have to deal with a list of references that point to object in the heap. (And if we do use struct, we don’t often create large struct since creating struct that has large size is a bad practice in C#)

CPU cache plays a very important role in performance improvement of programs on modern computers.

References

• [1] vector vs list performance benchmark, Baptiste Wicht: http://baptiste-wicht.com/posts/2012/11/cpp-benchmark-vector-vs-list.html
• [2] C++ vector, list, deque performance benchmark, Baptiste Wicht: http://baptiste-wicht.com/posts/2012/12/cpp-benchmark-vector-list-deque.html
• [3] Locality of reference: https://en.wikipedia.org/wiki/Locality_of_reference
• [4] Understanding CPU caching and performance, Jon Stokes: http://arstechnica.com/gadgets/2002/07/caching/

I’m using CentOS 6.2 64 bit, so I will download Instant Client for Linux x86-64, there any many packages, however, only two package we need:

1. Instant Client Package – Basic: All files required to run OCI, OCCI, and JDBC-OCI applications.
2. *Instant Client Package – Basic Lite: Smaller version of the Basic, with only English error messages and Unicode, ASCII, and Western European character set support.
3. *Instant Client Package – SDK: Additional header files and an example makefile for developing Oracle applications with Instant Client.

The first and second is the same, Basic Lite is smaller than Basic with fewer language support. Everyone need to choose between one of them. In my opinion, I install the first and the last packages.

• Instant Client Package – Basic for Linux 64 bit provide share library at /usr/lib/oracle/11.2/client64/lib, this is required libraries to run OCCI program.
• Instance Client Package – SDK for Linux 64 bit provide additional header files at /usr/include/oracle/11.2/client64, this is c++ header files that we need to include in our program.

Next step, we must tell Eclipse the path to the share libraries and the header files by: on Eclipse top menu, go to Project/Properties/C++ General/Path and Symbols

Eclipse Path and Symbols properties.

To-do list:

• Add header files directory into Include tab.
  (Add /usr/include/oracle/11.2/client64 into GNU C++)
• Add libraries directory into Library Path tab.
  (Add /usr/lib/oracle/11.2/client64/lib)
• Tell Eclipse which library that you want to link with your application by add needed libraries name into Libraries tab.
  In my situation, I need to add nnz11 (libnnz11.so in libraries directory), clntsh (libclntsh.so), occi (libocci.so).

Required OCCI libraries

We need only occi library but nnz11 and clntsh is required by occi, so we need to add them first. If not, Console may be throw an error the same as:

Additional, we need to set enviroment LD_LIBRARY_PATH = /usr/lib/oracle/11.2/client64/lib. It’s important to let our library know another. We can set this environment by append two line into ~/.bash_profile file (you may need to logout and relogin after changing this file):

Remember, libraries is require to run our application and header file is require for development. If we have already finish all steps, we can create a demo project that connect to oracle database.

If build successfully, we can sure that adding libraries and header step are ok. This demo project will execute query “SELECT * FROOM my_table”.

Hope that help!

Begining, we create our extension folder structure, if my company named Bb, and my extension name is Product, I will create the folder structure /app/code/community/Bb/Product. It’s mean our extension stay in community Code Pool and our extension is Bb_Product.

Store your extension information in /app/code/community/Bb/Product/etc/config.xml. The simple extension config file may be follow:

Now, our extension has it own profile, to let Magento load and control your extension, you must tell it “Heyx, load my extension please, my extension is Bb_Product in community Code Pool!”, so just create /app/code/etc/modules/Bb_Product.xml file:

That enough, our extension now can be loaded by Magento and you can enable or disable our extension in Admin Control Panel at System/Configuration/Advanced.

The next mission is create a page that output “Hello world”. Magento is base on Zend, so Magento is MVC base. If you don’t know anything about MVC, you should try to understand it first!

Create IndexControler.php file in /app/code/community/Bb/Product/controllers (offcourse, create /app/code/community/Bb/Product/controllers first)

Tell Magento know when should it use our controller, append some code into /app/code/community/Bb/Product/etc/config.xml.

You have already created Hello world magento extension. Visit our extension at:

• http://yourdomain.com/index.php/bbproduct/index/index (it is the same with http://yourdomain.com/index.php/bbproduct/index or http://yourdomain.com/index.php/bbproduct).
  You should see “Hello world” text.
• http://yourdomain.com/index.php/bbproduct/index/myaction.
  This output “Hello world, this is myaction”
• http://yourdomain.com/index.php/bbproduct/index/myaction2
  This action will output default Magento page template.