Multi-core processor - CPU, containing two or more computing cores on one processor chip or in one package.
Among multi-core processors at this point we can highlight
*processors designed primarily for embedded and mobile applications, in which much attention was paid by developers to means and methods for reducing energy consumption (SEAforth (SEAforth24, seaforth40), Tile (Tile36, Tile64, Tile64pro), AsAP-II, CSX700);
*processors for computing or graphics stations, where energy consumption issues are not so critical (graphical processors, for example, g80 series processors from NVIDIA, the Larrabee project from Intel, this also includes the Cell processor from IBM, although the number of processing cores is relatively low) ;
* so-called processors mainstream - intended for servers, workstations and personal computers (AMD, Intel, Sun);
- Number of cores (Number of cores. Core) - a silicon crystal with an area of approximately one square centimeter, on which microscopic logic elements are used to implement circuit diagram processor, so-called architecture. Each core is perceived by the system as a separate, independent processor, with all the necessary set of functions.)
Clock frequency (a clock is an elementary operation per second that a processor can perform. Therefore, the number of clock cycles is an indicator of how many operations per second a processor can process. The unit of measurement for this parameter is gigahertz GHz.)
Cache memory (memory directly built into the processor and used to store and access frequently used data is called cache memory. It is divided into several levels - L1, L2 and L3. The higher level of cache memory has a larger volume, but less high-speed data access.)
Bit capacity (determines the amount of information that can be exchanged between the processor and RAM in one clock cycle. This parameter is measured in bits. The capacity parameter affects the amount of possible random access memory- A 32-bit processor can only work with 4 GB of RAM.)
Performance
Power consumption
Dimensions
Price
Classes of tasks for which they are designed
Comparative characteristics processor performance, power consumption and data exchange speeds are presented in the tables
(Mflops - million floating point operations per second)
A significant contribution to the overall performance of the processor and the efficiency of its operation is made by the structure of inter-core connections and the organization of the memory subsystem, in particular cache memory
Processor CSX700
The CSX700 processor architecture was designed to address the so-called Size, Weight and Power (SWAP) problem that typically plagues embedded high-performance applications. By integrating processors, system interfaces and built-in error-correcting memory, the CSX700 is a cost-effective, reliable, and performant solution that meets the demands of today's applications.
The processor architecture is optimized for massive data parallelism and is designed with high degree efficiency and reliability. The architecture is aimed at intelligent signal processing and image processing in the time and frequency domains.
The CSX700 chip contains 192 high-performance processor cores, integrated buffer memory 256 KB in size (two banks of 128 KB each), data cache and instruction cache, ECC protection of internal and external memory, built-in direct memory access controller. ClearConnect NoC technology is used to provide on-chip and inter-chip networks (Fig. 11).
The processor consists of two relatively independent MTAP (MultiThreaded Array Processor) modules containing instruction and data caches, control units for processor elements, and a set of 96 computing cores (Fig. 12).
Rice. 12. MTAP block structure
Each core has a dual floating-point unit (addition, multiplication, division, square root, single and double precision numbers supported), 6 KB of high-performance RAM, and a 128-byte register file. 64-bit virtual address space and 48-bit real address space are supported.
Specifications processor:
core clock frequency 250 MHz;
96 GFlops for double or single precision data;
supports 75 GFlops for double precision matrix multiplication (DGEMM) benchmark;
performance of integer operations 48 ШАОs;
power dissipation 9 W;
throughput internal memory buses 192 GB/s;
two external memory buses 4 GB/s;
data exchange speed between individual processors 4 GB/s;
interfaces PCIe, 2 DDR2 DRAM (64 bits).
Designed for low power systems, this processor runs at a relatively low clock speed and has a frequency control mechanism that allows you to adjust application performance to specific power consumption and thermal environments.
The CSX700 is supported by a professional development environment (SDK) based on Eclipse technology with visual application debugging tools based on an optimized ANSI C compiler with extensions for parallel programming. In addition to the standard C library, there is a set of optimized libraries with functions such as FFT, BLAS, LAPACK, etc.
Modern Intel and AMD processors
The modern processor market is divided by two main competitors – Intel and AMD.
Processors from Intel, today are considered the most productive, thanks to the Core i7 Extreme Edition family. Depending on the model, they can have up to 6 cores simultaneously, clock speeds of up to 3300 MHz and up to 15 MB of L3 cache. The most popular cores in the desktop processor segment are based on Intel - Ivy Bridge and Sandy Bridge.
Intel processors use proprietary proprietary technologies to improve system efficiency.
1. Hyper Threading- Due to this technology, each physical core of the processor is capable of processing two threads of calculations simultaneously, it turns out that the number of logical cores actually doubles.
2. Turbo Boost - Allows the user to automatically overclock the processor without exceeding the maximum permissible core operating temperature limit.
3. Intel QuickPath Interconnect (QPI) - The QPI ring bus connects all processor components, thereby minimizing all possible delays in information exchange.
4. Visualization Technology - Hardware support for virtualization solutions.
5. Intel Execute Disable Bit - Practically antivirus program, it provides hardware protection against possible virus attacks based on buffer overflow technology.
6. Intel SpeedStep - a tool that allows you to change the voltage and frequency levels depending on the load created on the processor.
Core i7 – on this moment top line of the company
Core i5 - high performance
Core i3 – low price, high/medium performance
The most fast processors AMD is still slower than the fastest Intel processors (data as of November 2010). But thanks to my good ratio prices and quality, AMD processors, mainly for desktop PCs, are an excellent alternative to Intel processors.
For Athlon II and Phenom II processors, not only the clock speed is important, but also the number of processor cores. Athlon II and Phenom II, depending on the model, can have two three or four cores. Six-core model – Highend Phenom II series only.
Majority modern processors created by AMD support the following technologies by default:
1. AMD Turbo CORE - This technology is designed to automatically regulate the performance of all processor cores through controlled overclocking (a similar technology from Intel is called TurboBoost).
2. AVX (Advanced Vector Extensions), XOP and FMA4 - A tool that has an extended set of commands specifically designed for working with floating point numbers. Definitely a useful toolkit.
3. AES (Advanced Encryption Standard) - B software applications using data encryption improves performance.
4. AMD Visualization (AMD-V) - This virtualization technology helps ensure the sharing of the resources of one computer between several virtual machines.
5. AMD PowcrNow! - Power management technology. They help the user achieve improved performance by dynamically activating and deactivating parts of the processor.
6. NX Bit - A unique anti-virus technology that helps prevent infection of a personal computer by certain types of malware.
Use in GIS
Geographic information systems are multifunctional tools for analyzing consolidated tabular, text and cartographic data, demographic, statistical, land, municipal, address and other information. A lot of nuclear processors are necessary for quickly processing various types of information, as they significantly speed up and distribute the work of programs.
CONCLUSION
The move to multi-core processors is becoming a major focus for performance improvements. At the moment, 4 and 6 core processors are considered the most common. Each core is perceived by the system as a separate, independent processor, with all the necessary set of functions. The technology of multi-core processors has made it possible to parallelize calculation operations, as a result of which the performance of the PC has increased.
http://www.intuit.ru/department/hardware/mcoreproc/15/
http://kit-e.ru/articles/build_in_systems/2010_2_92.php
http://softrew.ru/instructions/266-sovremennye-processory.html
http://it-notes.info/centralnyj-processor/
http://www.mediamarkt.ru/mp/article/AMD,847020.html
Benefits of multi-core processors
The ability to distribute the work of programs, for example, main application tasks and background tasks operating system, across multiple cores;
Increasing the speed of programs;
Computation-intensive processes run much faster;
More efficient use of computationally intensive multimedia applications (for example, video editors);
Reduced energy consumption;
The PC user's work becomes more comfortable;
Nowadays, the minimum acceptable standard for equipping more or less serious computing equipment is considered to be the presence of a dual-core processor. Moreover, this parameter relevant even for mobile computing devices, tablet PCs and reputable smartphone communicators. Therefore, we will figure out what kind of kernels these are and why it is important for any user to know about them.
The essence in simple words
The first dual-core chip, intended specifically for mass consumption, appeared in May 2005. The product was called Pentium D (formally belonged to the Pentium 4 series). Previously, similar structural solutions were used on servers and for specific purposes; they were not inserted into personal computers.
In general, the processor itself (microprocessor, CPU, Central Processing Unit, central processing unit, CPU) is a crystal onto which billions of microscopic transistors, resistors and conductors are applied using nanotechnology. Then gold contacts are sprayed, the “pebble” is mounted in the chip body, and then all this is integrated into the chipset.
Now imagine that two such crystals are installed inside the microcircuit. On the same substrate, interconnected and acting as a single device. This is the two-core subject of discussion.
Of course, two “pebbles” are not the limit. At the time of writing, a PC equipped with a chip with four cores is considered powerful, not counting the computing resources of the video card. Well, thanks to the efforts of AMD, servers already use as many as sixteen.
Terminology nuances
Each die typically has its own L1 cache. However, if they have a common second-level one, then it is still one microprocessor, and not two (or more) independent ones.
A core can be called a full-fledged separate processor only if it has its own cache of both levels. But this is only necessary for use on very powerful servers and all kinds of supercomputers (favorite toys of scientists).
However, the “Task Manager” in Windows OS or the “System Monitor” in GNU/Linux can show kernels as CPUs. I mean, CPU 1 (CPU 1), CPU 2 (CPU 2) and so on. Don’t let this mislead you, because the program’s duty is not to understand the engineering and architectural nuances, but only to interactively display the loading of each of the crystals.
This means that we smoothly move on to this very loading and, in general, to questions of the expediency of the phenomenon as such.
Why is this necessary?
A number of cores different from one is intended primarily for parallelizing the tasks being performed.
Let's say you turn on your laptop and are reading sites on the World Wide Web. Scripts with which modern web pages are simply obscenely overloaded (except for mobile versions) will be processed by only one core. One hundred percent of the load will fall on it if something bad drives the browser crazy.
The second crystal will continue to operate in normal mode and will allow you to cope with the situation - at a minimum, open “System Monitor” (or a terminal emulator) and forcefully terminate the crazy program.
By the way, it is in the “System Monitor” that you can see with your own eyes which software has suddenly gone off the rails and which of the “pebbles” is causing the cooler to howl desperately.
Some programs are initially optimized for multi-core processor architecture and immediately send different data streams to different crystals. Well, ordinary applications are processed according to the “one thread - one core” principle.
That is, the performance gain will be noticeable if more than one thread is running at the same time. Well, since almost all operating systems are multitasking, the positive effect of parallelization will appear almost constantly.
How to live with it
Regarding consumer computing technology, single-core chips today are mainly ARM processors in simple phones and miniature media players. Outstanding performance from such devices is not required. The maximum is to launch the Opera Mini browser, an ICQ client, a simple game, and other unpretentious applications in Java.
Everything else, starting even with the cheapest tablets, must have at least two crystals in the chip, as stated in the preamble. Buy these things. Based at least on the considerations that almost all user software is rapidly getting fatter, consuming more and more system resources, so the power reserve won't hurt at all.
Previous publications:
What is the advantage dual core processors?
When buying a laptop, you probably noticed that some of them have labels: " Intel Core 2 Duo" or "AMD Turion 64 x2". These labels indicate that the laptops are built with dual-core processing technology.
Dual-core processors
Dual-core processors refer to a type of system consisting of two independent processor cores combined into one integrated circuit (IC) or, as professionals say, into a single chip. Such systems combine two cores in one processor. A similar technology was first applied to personal computer and to the home gaming console, but very soon it was adapted to the mobile computer environment. AMD and Intel have laptops with similar technology.
Dual-core processors have a different structure than dual single-core processors. They refer to a system where two processors are combined in one integrated circuit. Dual single-core processors, in turn, refer to a system where two independent processors (each with its own matrix) are directly connected to the motherboard.
Each of the processors in a dual-core system has an on-chip cache (primary cache), giving them the inherent potential to quickly and efficiently retrieve and process frequently used instructions. In addition, the same integrated circuit houses the L2 cache. The secondary cache memory on Intel's Mobile Core 2 Duo chipset is shared between two processors. In the Turion AMD 64x2 chipset, each of the two processors has a dedicated cache memory - 512 KB per core. The second level cache is a reserve for case if the primary one is not enough.
Benefits of Dual Core Technology
The most important advantages of such processors are speed and efficiency. Command processing and data retrieval are carried out by two processors; thus, greater performance is achieved without heating the processors. The fact that these two processors have their own easily accessible primary cache also ensures fast performance. Additionally, especially in the case of the Intel Core 2 Duo, where the secondary cache is split, the entire secondary cache can be used by either or both processors simultaneously if the need arises.
In a nutshell, a laptop with a dual-core processor runs faster, runs cooler, and has better multitasking capabilities. Dual-core processors consume less power than dual single-core processors.
Another advantage of using dual-core processors in laptops is their lighter weight and size, making the laptop more convenient while delivering PC-like performance.
It's important to note that with older programs, if you only run one program at a time, you won't experience any benefit from dual-core processors. Older programs were not designed for this technology, so they are only able to use one core. However, in this case, the advantage of multitasking still remains. If you have multiple programs open at the same time, a dual-core processor will provide more fast performance than single core.
Time goes by, and more and more developers software create their programs taking into account dual-core processors; Thus, users will be able to experience all the benefits of such processors in the near future.
A dual-core computer is a computer whose central processing unit has two cores. This technology makes it possible to increase the productivity of its work to a fairly large extent.
What is a dual-core processor?
A dual-core processor is a processor that has two cores on one chip. Each of the cores usually has a Net Burst architecture. Some of the dual-core processors also support Hyper-Threading technology. This technology allows processes to be processed in four independent threads. This means that one such dual-core processor with this technology (physical) replaces or is equivalent to four logical processors, from the point of view of the operating system.
So, each core of a dual-core processor has its own L2 cache of a certain amount of memory, as well as a shared cache with twice as much memory. As a rule, the crystals on which dual-core processors are manufactured have a size of about two hundred square millimeters with a number of transistors exceeding two hundred million units. It is worth noting that with such a huge number of elements, this processor, it would seem, should highlight a large number of heat and therefore cool down accordingly. However, it is not.
The highest crystal surface temperature is about 70°C. This is due to the fact that the voltage supplying the processor does not exceed one and a half volts, and highest value The current is one hundred twenty-five amperes. Thus, increasing the number of cores does not lead to a significant increase in power consumption, which is very important.
Advantages of computers with dual-core processors
The need to increase the number of processor cores arose when it became clear that further increase in its clock frequency does not lead to significant improvements in performance. Computers with dual-core processors are aimed at using applications that use multi-threaded information processing. Therefore, the benefit of such a computer is not possible for all programs. Programs that use the capabilities of two cores include, for example, programs for rendering three-dimensional scenes, programs for processing video images or audio data. Also, a dual-core processor will be beneficial when running several programs on a PC simultaneously. In this regard, such processors are usually used in computers designed to work with graphics, as well as to work with office programs. Thus, for gaming needs this technology the second core is almost useless.