Cerf: the future of the Internet does not include Issue 7 (1)

BKJIA, Internet headlines in February 22] Vint Cerf is Google's chief Internet consultant. He is currently working on several projects related to the next-generation Internet. One of these projects is to promote the adoption of IPv6 globally, but the plan for IPv7 is not on the agenda.

To this end, at the annual Electronic broadband Migration Conference held in Boulder, Colorado, Julie Bort, Cisco editor of Network World, held a long talk with Cerf, this article discusses the future of IP addresses, home networks, the Internet of Things (IOT), the so-called "damage toggle" of the internet, and other issues.

Will 7 be displayed in the future? What new problems will it solve?

When trying to find out how to deal with the address space expansion problem, we actually sum up four suggestions. These suggestions are eventually reduced to one. So in fact, 7, 8, and 9 won't appear. There does not seem to be any new ideas. The most critical issue now is how to make IPv6 widely available before we gradually exhaust IPv4 address space.

However, we should not prevent anyone from thinking about re-designing the internet. In fact, a topic named "innocent history" supported by the National Foundation for scientists is being conducted at Stanford University. They are trying to see how we design today's Internet, let it cover everything we know. So far, at least one specific achievement has been "openflow Router", which can process data packets in a completely different way from traditional routers. But as far as I know, there is no new modification plan for the IP protocol. But this day will always come, and I believe it.

In the next ten years, do public/general consumers need to ensure that all their home equipment, routers, and so on have entered IPv6, or is the efforts of NAT and operators to bridge the gap between IPv4 and IPv6?

No, this is not enough. I am not a loyal fan of Carrier-level network address translation. This is partly because the concept of network address translation is very fragile and does not allow servers to be put into use on the premise of consumers.

When the Internet is highly asymmetrical, your download speed will be faster than the upload speed. The practice of placing a server at home is not popular with broadband vendors. Because it consumes a lot of uplink bandwidth capacity. However, over time, I believe that this is not only desirable but also a natural thing. People will own servers or even use the cloud at home. Therefore, symmetric capacity and IPv6 will be very attractive. I said that I am not a supporter of Carrier-level network address translation-but because we are running two Protocols at the same time, therefore, network address translation is necessary to achieve excessive traffic during this period.

Do you think it is as urgent to implement IPv6 at home as it is in business networks?

I think it is very urgent, because if one day we cannot run two Protocols at the same time, then some servers can only run IPv6, or some users may be able to run only IPv6, but others will be helpless, because NAT has been exhausted.

I think the most important thing is to allow both protocols to run smoothly at home. The current Laptops and Desktops already have this capability. Generally, the firewall, NAT host, or broadband modem you use at home are not ready for IPv6 Configuration.

So on September 6, June 8, Global IPv6) when we were conducting a 24-hour global test of turning to IPv6, I believe there must be some problems that need to be solved one by one.

I hope to see a concerted effort to implement IPv6 together. Then it will enter a stable and excessive stage. In fact, running IPv4 or IPv6. what matters is that communication between people is free of barriers.