In fact, it's been two days already, but no one has yet written an article on Habr, so I will have to eliminate this omission, which I do with pleasure.
So, what's new in this version of PostgreSQL?
First, the versioning itself has changed. Prior to the "dozens", we observed a lot of minor versions of 9.x that came out about once a year and at the same time introduced serious, far from minor changes. Therefore, since version 10, it has been decided to number 10, 11, 12, etc. By the way, MySQL seems to have gone the same way, jumped from 5.7 to 8.0
Okay, it's all trivia, let's move on to the essence of the question
China is funding the development of the world's first prototype supercomputer for exaflop computing. The speed of the system will be a quintillion operations per second. Such a capacity the country will send to support the program of marine research.
Improving performance of the disk subsystem is the most topical issue today. This is caused by the low cost of HDD that takes a leading position in the mass segment. More than a half of the resource-intensive applications have a "vulnerable spot" that are the spindle drives. In this case, everything does not depend on the bandwidth of SATA interface, but it depends on the physical capabilities of the mechanical components of the magnetic disk. The bandwidth of SATA-II and SATA-III interfaces makes 300MB/s and 600MB/s respectively, and the maximum performance that can provide a regular HDD does not exceed 150MB/s. Therefore, the transition to SATA-III-interface will be reasonable only for SSD, but it is not suitable for everyone.
In order to evaluate and to test the effectiveness of different types of disk subsystems were chosen following solutions:
1. OCZ RevoDrive X2 PCI-E SSD 100Gb
2. Plextor PX-128M2S SSD 128Gb
Each of us complained about the forecasters, when finds to be in the rain without an umbrella. In this article, we will tell you how to make a forecast based on the model WRF (Weather Research & Forecasting Model) for the several days in advance.
A few words about WRF
WRF is a numerical weather prediction model, which is suitable to predict the condition of the atmosphere, as well for research. It has been developed by the community of scientific organizations the United States, including the National Center for Atmospheric and Ocean, the National Center for Atmospheric Research.
It is a system of modules: the preparation module of initial and interfacial data (WRF Preprocessing System), it is the actual determinant core (Advanced Research WRF), the post-processing module (WRF Postprocessing System).
Most part of the model is implemented in the language fortran, which is using the library MPI. Modules are written on C for working with data. The model is available in the source codes.
Naturally, the model consumes an enormous amount of CPU time, and it is designed to run on supercomputers, since the weather forecast is one of the toughest tasks.
High-loaded project (website) is not necessarily a popular social net, video hosting or MMORPG. The easiest way deeply to improve the website’s requirements for the hardware is transferring storage of sessions in the database. In this article we will discuss how to store data in the database and at the same time does not cut on the performance. Using a small amount of RAM we can quite well save the time of CPU. We're talking about situation when memcached is not available and other special resources of caching.
Magic MEMORY tables
DBMS MySQL database system realizes the type of tables that are stored in memory permanently and always are available for a short time. This is MEMORY, there is a synonym HEAP. The second name is older; therefore, it is preferable to use the first one.
Comparing with MyISAM or InnoDB, this format is a very limited, but it handles well the storage of operational data. Traditionally, we will give its pluses and minuses, we start with the pluses: