Having researched the history of the Internet and networking in general, I have a solid idea of how we arrived at the complex, vulnerable and expensive networking architecture we have today. Now I want to examine the fundamental flaw that has resided in TCP/IP since the protocol was first proposed in 1974. The use of the IP address for both location and identity.
First, we need to be clear that TCP/IP is not going away anytime soon. IPv4 may be running out of addresses, but it still provides the foundation for virtually every network in the world. This includes both the public internet and private corporate, government and personal networks. Even with the introduction of the more robust IPv6, the fundamental principles of TCP/IP apply.
The principle behind TCP/IP is to be able to easily and reliably connect everything. It’s better in concept than in actual application. When you connect everything, with a flawed model, you are inviting trouble. Hackers, cyber-criminals and government sponsored electronic espionage have proliferated as the Internet has grown and as almost every system on earth has become connected.
As I mentioned, the fundamental flaw within TCP/IP is that it combines location and identity in a single address, using what we refer to it at Tempered Networks as an address defined networking model. This creates an inherent openness across all networked things, consequently resulting in a lack of security. Simply put, the security problem is caused by TCP/IP’s use of a connected device’s address to serve the dual purpose of identifying the device, as well as providing its location on the network. This results in a network vulnerability that is very visible to, and easily spoofed by, hackers anywhere in the world. And with identity also being used as a device’s address, hackers can often fake a valid IP address to gain access into your network. Once inside, they can steal data, disrupt service, and generally wreak large-scale havoc.
We hear about it all the time too. Think about the last time you read a story covering a massive network breach or a company dealing with a ransomware attack. Chances are it was within the past week. These attacks can be a financial disaster for a company or lead to personal ruin for individuals. With this in mind, the question is, do you really feel comfortable relying on your IT team to wield overall responsibility for keeping your network safe or quickly recovering from an attack? Probably not. I know I wouldn’t.
Blame TCP/IP. It is open, reliable and fast, which is what it was always meant to be, but the openness and way we use IP addresses make it extremely, inherently, vulnerable. Since every device has an IP address, it makes every other device or system it is connect with open to attack. Even worse, many devices have fixed IP addresses that can cause significant headaches when there are multiple devices on a network with the same one. This means that you not only have to secure the network from intrusion, but you must set very specific routing rules to connect two devices with the same IP address.
To secure these open, unsecure IP networks, companies implement firewalls, security protocols and IP routing schemes that are intended to keep devices within the network connected while keeping hackers and other unwanted visitors out. Unfortunately, this activity, combined with multiple IP address conflicts, frequently creates situations where it is impossible for a device on one network to talk to a device on another network. And with the recent push by many companies to the cloud, these conflicts and configuration requirements have only become worse.
All of this adds up to a level of complexity that is unsustainable. Networks are costly to maintain and manage, require significant expertise to configure and secure, and it can take weeks to move or provision a device that has a fixed IP address. Even making significant investments in IT infrastructure and personnel is no guarantee that your network is secure or can react to various needs at the speed the business requires to maintain efficiency. There must be a better way!
At Tempered Networks, we believe the future is in identity defined networking (IDN) based on the recently ratified Host Identity Protocol (HIP). HIP fixes the fundamental flaw in TCP/IP by moving from an address based, open trust networking model, to an identity based, zero trust networking model. It still uses IP addresses for location, but requires devices to recognize the specific, cryptographic identity of another allowed and approved device before any communication takes place. By doing this, it eliminates the security vulnerabilities inherent in TCP/IP while also removing IP conflict issues that present a constant struggle for IT departments.
Over the past 35+ years the world has become connected in ways that the inventors of TCP/IP never could have imagined. It’s time to move to a modern, identity defined networking standard and stop relying on flawed standards that are not suitable for a mobile world. A world in which everything needs to be securely networked no matter where, when or how it is connected.