Clouds are not everything

We are all delighted with the development of cloud computing and the advent of services like icloud. On the other hand, when we are keen to discuss the latest versions of Samsung Galaxy, HTC and iphone 4S, we tend not to move forward or backwards to think about the evolution of mobile devices and networks. This is particularly important because there is a view that cloud computing can compensate for the lack of performance on mobile devices, so our mobile devices need not be "too smart". Still, we have to focus on what the future of mobile devices will look like.

A quick review of storage device costs

The first laptop I used was Kaypro 10. It was 1983, the CPU frequency of this computer is only 2.5 MHz, hard disk 10MB. By 1993, I had a 250MB hard drive for my Dell desktop, 700 dollars. The third data was 1999, when I first bought a 16M CF card that cost me 80 dollars. This CF card is also the first mobile storage device in my real sense.

Today, our smartphone has a dual-core CPU, which is more than 1GHz, and we can extend the SDHC Flash memory card for our mobile device, with an average price of $1 per gigabyte. Now 700 dollars can buy a notebook with 512GB SSD, a 1TB (1TB=1000GB=1000000MB) hard drive is only 40 dollars.

In my example, as you can see, in the last 12 years, the mobile memory card from every 4800 dollars GB to 1 U.S. dollars GB, hard drive from 2.8 million dollars terabyte to 40 dollars terabyte. If the inflation factor is taken into account, today's costs will be lower.

Clouds are not everything

What does this have to do with cloud computing? The concept of "cloud is Everything" is a mobile device that represents our future use of our "real smartphone" to enjoy ubiquitous, wireless, low-cost connections, but this does not take into account the evolution of mobile networks and the capabilities of devices themselves.

For digital signal transmission, the first rule of Nyquist shows the relationship between baseband width and network code rate, which may become the bottleneck of cloud. The answer to this question is that we can use the device itself more.

The power of the cloud depends on the power of the device.

Imagine, in 2020, that our phone was using multi-core processors, up to a few terabytes of local storage, hundreds of GB of songs are stored in the phone (at that time people are more likely to have higher quality requirements, files are larger), several terabytes of movies, hundreds of GB of photos, plus the hottest 3D games, Save our favorite sites and social networking content to the local, well, 10TB space is gone.

Over the last few decades we have solved many of the problems of infinite networks, but on the WAN, we are getting closer to the limits of Shannon (note: Shannon limit, or Shannon capacity, refers to the maximum transmission rate of error-free transmission on a channel that will randomly error-prone), that is to say, in a limited frequency band, There is a limit to the transmission of data. How do you ensure seamless transmission of these hundreds of GB or even terabytes of data streams? We need to take into account real-time transmission, bandwidth requirements, different forms of content loading in the local network to arrive at one of the most economical solutions.

The reason I'm saying this is to want you to realize the nature of the cloud. As the saying goes, a long time must be combined, a long time must be divided, we have to take into account the limitations of the cloud, the wireless service, and especially the transmission device itself, and as the future of the mobile device itself is growing in mass storage management and computing power, we will find that instead of fully pinning our hopes on the cloud, more depends on the device in our hands.