Advantages and disadvantages of

advanced computing and cloud computing Advantages and disadvantages of advanced computing and cloud computing

The growing number of smart devices gives advanced computing an edge, but does this mean that conventional cloud data centers will become decentralized and the need for a centralized cloud disappear at the same time? future? The use of cloud services has grown rapidly because it allows companies to get more processing and storage power, while paying only when needed and at a lower cost instead of companies spending on IT equipment alone. On the other hand, cloud providers are looking for opportunities to optimize their investment. Processors and storage devices have increased in performance exponentially while getting smaller and smaller, a low-end smartphone is more powerful than a room-filling computer in the 80s. With more powerful devices in the hands of consumers (laptops, smartphones, etc.) and at the general edge (IoT smart devices), it is increasingly desirable to offload some of the computing activity onto these devices. This allows the cloud provider to reduce the need for computing resources and provide users a faster and more personalized experience because only a minimal amount of data is transferred over the network. Edge computing is useful for service providers because it uses the available capacity of users' devices instead of the power of cloud computing, and it speeds up the user experience because less data needs to be transmitted over the network. On the other hand, this does not guarantee sufficient trust or security, because the device can be manipulated by the user with little control from the point of view of the service provider.

What does the future of the cloud look like?

Simone Vernacchia, Partner and Head of Digital Security Infrastructure, Resilience and Cyber ​​Security at PwC Middle East, told TechRadar Middle East that cloud computing will not disappear with advanced computing gaining ground "Edge computing is very promising and at the same time, you have a lot of potential problems. Whenever you need real processing power, you prefer to use the cloud. Both technologies are somewhat complementary," he said. On the one hand, she said that if you want to use the processing power of the cloud, you have to transfer all the data to the cloud, wait for the processing to happen, and then retrieve the data. "This will introduce delays due to two different phenomena: the first is that the bandwidth is not infinite and can become saturated if we want to perform computations on a large amount of data, and the second is latency," she said. he stated. "In a real network, we can try to have a cloud provider closer, but we can't go faster than the speed of light (light is actually the way bits are transmitted on a physical level in high-speed networks). speed)," he said. When you put a network between your screen and the keyboard, he says that you must use a pipe for everything you do.

Will the latency reach the speed of light?

"Everything you do has to go through a geographic network. The geographic network has two things: bandwidth is limited because technology can't evolve beyond a certain gigabyte, and the second is latency, which means latency It can reach a maximum is the speed of light and no faster than that. We are far from the latency to reach the speed of light," Vernacchia said. In addition, he said that latency is a much more difficult problem to solve from a physical point of view, as well as from a protocol point of view, but it will get better and better. but it will not reach the speed of light. For any network traffic to be transmitted, there are different layers of protocols. A protocol is a set of rules that organizes the information in packets and controls that they are not affected by the errors introduced by the transmission through the network. Vernacchia said the protocols consume data just to shape the packets and make sure they arrive complete or are corrected once they arrive. "This overhead introduces additional delay because it consumes bandwidth." To reduce the amount of data they need to transport from one place to another, service providers are trying to transfer more and more computations to cutting-edge devices, saving costs and money. offering a faster, smoother user experience," he said. . . . "As more and more users use the network, we need more bandwidth and latency. Edge computing is trying to solve this problem by doing a lot of computation on your device to reduce latency," he said. The edge advantage, he says, is that you pay less to a cloud provider, but the problem is that the edge computing chip Your device may not be powerful enough to do it all and you may have to rely on the cloud because the power and storage you can get is much more extensive than on the on-board device, and since the cloud provider has no control about your device, the result may not be reliable if you manipulate the device.

Are data centers needed in each country?

Vernacchia replied yes, or at least he wants to have dedicated entry points for broadband network connections close to his country to minimize latency and maximize bandwidth. "In addition, cloud providers will move digital data as much as possible to be close to you in order to optimize and invisibly (and out of control) to the user. This is called "charging" to provide the lowest latency. short available. In other cases, cloud providers will (quietly) move user data from one country to another where they have more or less storage," he said. Going forward, he said cloud providers will try to push as much as possible to reduce costs and where security is not critical. "Edge computing is now reaching its limits and consumers are enjoying a better experience," she said. For example, in the case of heavy video games that have a lot of visual effects, in order to reduce the required bandwidth and improve the user experience, processing is done more and more. on the device instead of in the cloud. On the other hand, to preserve security, when the user wants to make an in-app purchase, this is configured in the cloud.

Safety is an issue on the cable

"What advanced computing uses in applications that require low security requirements and what the cloud uses are things that need to be controlled and that require high security requirements." Vernacchia said. However, she said security is an issue at the edge because the device is owned and operated by the user, who can manipulate it without service providers being able to. avoid. "If you jailbreak your smartphone, even though it may violate some license agreements, it's very difficult to impose sanctions. On the other hand, if you go into a cloud provider, going after it is relatively more feasible," she said. "When you want performance, that's one possible way to see if a cloud provider has a data center in your country, but for most companies, performance is just a matter of time." part of the story For others, latency is the other part and for others, it's data security and compliance. Some customers would like to have a data center in Europe than in the United States, because the privacy rules are more protected," she said.