Blooming America

Seymour Clay, the father of supercomputers, was tossing and turning in the laboratory at this time. After finding that he could not sleep, Seymour Clay got up again at three o'clock in the middle of the night, and then looked at his design plan.

"Why is it still not working?"

The new generation of supercomputers will use Capricorn as the core.

When Seymour Cray got this kind of 32-bit microprocessor, he was taken over by this magical processor, using the latest technology, using the best architecture, and possessing ten times the size of the series. can……

And these, when Seymour Clay was designing a new generation of supercomputers, he was inspired.

At this time, Great God Clay suddenly thought of the new supercomputer architecture he had discussed with Catherine, and the combination of this architecture and the current new processor seemed to be a good decision!

Thinking of this, Clay immediately organized the manpower and started the research and development of the supercomputer.

But in the process of research, Clay was hindered by unprecedented obstacles.

The new supercomputer adopts a turbo-type design, and each blade is a server, which can realize collaborative work and can be increased or decreased at any time according to application needs.

This form of blade server is what Cray is preparing to use.

A blade server refers to a server unit that can be plugged into multiple card-type server units in a rack-mounted chassis of standard height. It is a low-cost server platform that implements hahd. This kind of server specially designed for high-density computing environments is more efficient than It turned out to be higher.

A blade server is like a "blade", each "blade" is actually a system motherboard.

These blade servers can start their own operating systems through the "onboard" hard disks, such as x system, ud system, etc., which are similar to individual servers. In this mode, each motherboard runs its own system and service The different user groups specified are not related to each other.

However, administrators can use system software to assemble these motherboards into a server cluster. In the cluster mode, all motherboards can be connected to provide a high-speed network environment and share resources at the same time to serve the same user group. Inserting a new "blade" in the cluster can improve the overall performance. And since each "blade" is hot-swappable, the system can be easily replaced and maintenance time is reduced

In the previous generation of Galaxy supercomputers, this concept was used to add modules to the computer.

However, compared with the current new technology perfected by Cray, such technology is inadequate.

The new technology can more thoroughly and skillfully use the computing power of supercomputers and improve the efficiency of the machine.

At the beginning of the design, these blade servers have the characteristics of low power consumption, small space, and low stand-alone price. At the same time, it also inherits and promotes some technical indicators of the Galaxy series of servers, such as applying hot swap and redundancy to the blade servers. Among them, these designs meet the demand for server performance in dense computing environments; some also use built-in load balancing technology to effectively improve the stability of the server and the performance of the core network. Moreover, the blade server can save the space and cost of the server to the greatest extent, and provide users with flexible and convenient means of expansion and upgrade.

Even so, Cray encountered various problems in the design.

First of all, this is a brand new architecture.

Secondly, the 32-bit processor is also a brand new architecture.

But at this time, Clay's previous experience was useless. Because at this time, he was exposed to a completely new technology. Such technology is very difficult to rely on Clay alone or even a team.

Clay realized that he needed a large team, and then after a long period of research, it was possible to achieve his goals.

One year...no, maybe two or three years, it is possible to achieve my goal.

All the previous things have to be overturned and redesigned, so it is impossible to do without such a long time.

In the 21st century, blade servers have become the mainstream of high-capacity computing clusters. In the world's top 500 supercomputers, many new cluster systems have adopted blade architecture. Because the blade server can be used to greatly reduce the number of external cables required, the hidden dangers caused by cable connection failures can be greatly reduced, and the reliability of the system can be improved.

And when Catherine was discussing with Cray, the two parties were talking about supercomputers composed of such blade servers.

In terms of servers, Catherine's classification is similar to that of her previous life, which is entirely based on Catherine's picture.

According to the different appearance structure, servers are divided into three types: tower, rack, and blade servers.

——But in fact, there are only the first two, because the third has not yet appeared when Catherine proposed this concept. This only exists in Catherine's notebook and Kray's mind.

Tower servers are generally the most common. Its appearance and structure are similar to those of ordinary machines, but they are slightly larger, and there is no uniform standard for their dimensions.

The motherboard of a tower server has a strong expansion and a lot of slots, and a lot of space is often reserved inside the chassis of a tower server for redundant expansion of hard disks, power supplies, etc. This kind of server does not require additional equipment, does not require much space, and has good scalability, and the configuration can be very high, so the application range is very wide, and it can meet the common server application requirements.

This type of server is especially suitable for common entry-level and workgroup-level server applications, and the cost is relatively low. Xìng can meet the requirements of most small and medium-sized enterprises users. The current market demand space is still very large.

For example, the server of Catherine's current personal website basically uses a tower server.

Moreover, some small companies also purchase and use this kind of server. In fact, they don't have much demand for such a computer.

The demand for computers is not the higher the better, and the cost is also a very important reference object.

However, this type of server also has many limitations. When multiple servers need to be used to work at the same time to meet higher server application requirements, it is very unsuitable due to their relatively large individual size, large footprint, and inconvenient management.

The rack server is actually a product under industrial standardization. Its appearance is designed according to a unified standard, and it is used uniformly with the cabinet to meet the enterprise's server intensive deployment needs. The main function of rack servers is to save space. Because multiple servers can be installed in a cabinet, it not only takes up less space, but also facilitates unified management.

The width of the rack server is 19 inches, and the height is in inches (mm). There are usually 1u, 2u, 3u, 4u, 5u, and 7u standard servers.

If it is the 21st century, the popular ones are probably 1u and 2u cases, but they are different in Catherine's era. Because the motherboards are "silly, big, black and thick", the 3u and 4u cases are almost the people's choice, and in the eyes of many people, this is already "small".

The advantage of this kind of server is that it occupies a small space and is convenient for unified management, but due to internal space limitations, expansion is more limited.

This kind of server is mostly used by large enterprises with a large number of servers, and many companies use this type of server, but the server is delivered to a dedicated server hosting organization for hosting, especially the servers of many websites currently use this method .

If you want a better company, you are using supercomputers. The first-generation Celeron has not been produced by the company for a long time, and the second-generation Galaxy is the mainstream in the market, and orders have even been arranged. In 1975.

Even the top ten consortiums use such supercomputers at this time, but such supercomputers are not formal "servers", just because such supercomputers are powerful enough to be competent.

After all, at the beginning, what Katherine needed was stability, and the cloud system was a typical example of this. Of course, such a machine could be used as a server.

But speaking of it, this "orthodox" choice is in the system Seymour Clay is currently studying.

"How can we bring out the efficiency of the entire system?"

According to Catherine, such a system is very efficient, but Clay discovered that he was not the case during the research process.

Because the blade server is more dense than the "Galaxy module"-that is, a separate module of the Galaxy computer.

The core of the blade server is a 32-bit core with better craftsmanship, but because of space, the amount of heat emission is higher than that of the "Galaxy module".

It's not just this problem~www.wuxiaspot.com~ In the first design, Cray found that the architecture he designed can only have at most 50% of the energy he expected, and more of the energy can be wasted. Up.

"Come over..."

Clay could only think so.

"Perhaps... I should change my mind..."

The new core comes with its own graphics card, but my thinking is still stuck on the old computer. Why can't I think about it in a different way?

If you can make full use of the xìng of gpu, it should be better...

...

By the way, our new computer is equipped with a server case, 4u standard, gtx670 graphics card, old roll 5 full speed 60 frames...

Three shifts today, the first shift~

It seems to have eaten something bad, my stomach hurts...! .