When Matt got off the plane, Duke had no idea that the slightly melancholy blonde man was coming for him. Duke, who came back from the New Big Drive ceremony, had no sense of these undercurrents targeting him. Now he was almost devoted to the J-7 drone transformation project. Duke happened to have a little free time in other projects for this project with high expectations for Mr. Liu, so he devoted a lot of effort to do his best to do this.

Compared with Jingwei and Winglong, this transformation plan is very challenging. First of all, because the combat radius used by this drone plan far exceeds the current level of Jingwei and Winglong-class drones. According to the combat concept of the assembly, the combat radius of the modified unmanned j-7 fighter is at least several hundred kilometers. Duke needs to develop a brand new autonomous identification system for them to discover and lock in combat targets.

Because this combat radius has exceeded the limit of Duke's wireless communication technology, a large number of complex operations can no longer be performed by transmitting real-time data to the ground monitoring workstation, but must be completed independently on the j-7 drone. This real-time data transmission is no problem with Duke's wireless communication technology within a range of 200 to 300 kilometers, but if the distance is larger, there is nothing to do.

Although satellite communication can be transmitted to farther places, the bandwidth cannot support the needs of large cluster operations of j-7 drones, so this complex operation can only be completed on j-7 drones.

However, from the perspective of cost-effectiveness, it is not cost-effective to equip each j-7 drone with a high-performance workstation that can complete this computing task. The cost of this high-performance workstation alone is as high as one million yuan, and the configuration is not much different from that of a small machine. Due to the large operating power requirement, it is also necessary to have a maximum generator on the machine to operate. This will increase the carrying weight of the j-7 drone that is not large.

For Duke, this is certainly not an ideal option. Although this solution needs to be simplified in application implementation, it can be basically implemented by simply moving the ground monitoring station computing system of the Winghound system to the J-7 drone. However, this does not conform to Duke's identity as a technological genius. It is difficult to satisfy Duke himself if he creates such a bulky and uncreative thing.

Therefore, configuring high-performance workstations for j-7 drones is not a good solution.

Duke took out the functional requirements and combat needs provided by Xiao Qianguang, and carefully studied the combat methods that the assembly imagined in the future that these drones might adopt. Duke found that there was a very significant feature, that is, these J-7 drones rarely perform separate combat missions, and reconnaissance did not require them. Duke would create better specialized unmanned reconnaissance aircraft for reconnaissance. After decrypting and imitating the Global Hawk's materials, sensor systems and power devices, Duke believed that his R&D team would soon be able to create a better reconnaissance aircraft than the Global Hawk, and there was no need for clumsy guys like the J-7.

The biggest purpose of this j-7 drone is to use group fighting tactics, that is, to adopt large cluster combat tactics that rush to the top. It faces imaginary combat targets such as enemy high-value cluster targets, such as airports, heavy weapon positions, and even large fleets.

In this combat mode, Duke felt that cluster computing could be used. Just like he had previously used supercomputer functions to achieve similar supercomputer functions through cluster computing, it connected multiple adjacent j-7 drones through wireless broadband, preparing to allow future j-7 drones to have cluster computing capabilities.

In this case, in order to adapt to the group fighting tactics of J-7 drones in the future, Duke did not need to choose to install 8 latest CPU super workstations to do this computing core. Although this workstation has a strong computing power, the price is 4 to 5 times higher than that of 4 similar workstation servers, but the computing power has not increased to such many multiples of the price.

If you choose 4 CPU workstation servers with lower cost and use more than 3 workstations to jointly operate, you can obtain the computing speed of approximately 8 CPU high-end super workstation servers.

According to the simulation tests on Kerry, the combined computing speed of two low-end workstations can basically be close to more than 80% of the original high-performance workstations. When 3 workstations are connected, it can exceed it. The larger the workstation cluster connected at the same time, the better the computing performance. Although this performance expansion is not unlimited, because on the one hand, the wireless transmission speed is limited, this joint cluster operation requires a large amount of data exchange. When the scale of the machine cluster expands to a certain extent, it will face this data transmission bottleneck.

On the other hand, the more servers you access at the same time, it actually means that more images will need to be processed at the same time. When several aircraft are flying adjacent to each other, a large part of the images they take are the same. The image processing application developed by Kerry can automatically recognize the same images, greatly speeding up the recognition of the same image "seeing" by adjacent aircraft. However, when clusters are dispatched, the images captured by aircraft far apart do not have so much similarity, so the processing will not be too fast.

Based on this feature, the new cluster computing application developed by Kerry can form a small computing cluster based on the coordinate position information of the aircraft cluster space through the wireless communication system, which can perfectly process the images passing through the region in real time, identify possible potential attack targets in the region to share with the entire cluster, or transmit the result back to the rear for confirmation through satellites.

This adds countless complexity to the software architecture. Without supercomputer assistance like Kerry, Duke would definitely not be able to complete these super complex cluster algorithms. However, after such a struggle, the cost of each J-7 drone spent on the workstation server was reduced to one quarter to one fifth of the original, which means that the renovation cost of a single J-7 was immediately reduced by hundreds of thousands of yuan. Considering that the future transformation is calculated at a scale of more than a thousand, Duke's change directly saved more than hundreds of millions of yuan.

Technology is the primary productive force fully reflected in this research and development process, and there is no loss in performance and function. Of course, this is a model for cluster combat. If a stand-alone combat is dispatched, there is still a huge difference. This once again shows how big the value difference between the human brain and the intelligent brain of Kerry.

(Special thanks to the book friend "lanxue" for supporting 6666 coins. This is the biggest reward this book has received. At the same time, I also thank the book friend "Who am I 4004" for supporting you. I can fully feel your enthusiasm. Today, I will add a special chapter to the book friend "lanxue" at 23:58 to congratulate Xiaofei!)
Chapter completed!