IPsec (Internet Protocol Security) and TLS (Transport Layer Security) are both used to protect data in transit between devices, and to secure network traffic over public or private internet. Both have cryptographic mechanisms for encryption, integrity checking, mutual authentication, replay protection, key exchange, etc. However, they operate at different network layers, and have different strengths and security goals in mind. The purpose of this blog post is to help our readers understand the differences.
IPsec
IPsec operates at the IP (Internet Protocol) network layer, and has reserved IP protocol numbers of 50 for ESP (Encapsulating Security Payload) and 51 for AH (Authentication Header) security protocols. AH provides integrity checks and data origin authentication, while ESP provides confidentiality, integrity checks, and authentication. IPsec supports both IPv4 and IPv6 protocol versions.
At the most basic level IPsec protects traffic between a pair of src and dst IP addresses that share security associations (SA). For example, if Alice and Bob use ip xfrm to provision shared SAs between their linux computers with ESP rfc4106(gcm(aes)), then all IP traffic between their two computers would be protected with authenticated encryption, regardless of whether they were carrying TCP or UDP, or some higher level application protocol. Thus IPsec provides blanket security guarantees that can be useful for large organizations and for constructing private networks. Furthermore, legacy applications that don’t support modern TLS can be protected.
The disadvantages of IPsec are that it can be complicated to configure and maintain, and in the embedded space the network stack may not support IPsec at all.
TLS
In contrast, TLS is agnostic of the underlying transport mechanism, and is not tied to a specific network abstraction layer. While the most common use case is over TCP (TLS) and UDP (DTLS), TLS can be implemented over UART or I2C in embedded devices. Therefore TLS is really about securing application layer communication: e.g. protecting data in transit between a web browser and an email server, or between remote sensors in the field and a hub collecting data.
The value of TLS is in its individualized and flexible nature. A user of TLS can know that their communication is protected even over an untrusted network. Furthermore, TLS can be used on bare metal devices that lack an operating system and only have a minimal network stack. Finally, TLS is the true standard of the bulk of public internet communications. As such, it is the most widely tested security protocol available.
IPsec vs TLS, and Zero Trust
In short, TLS and IPsec are simply two different models for solving the same problem, with unique strengths and tradeoffs. If you need to secure a device in the field that connects back to servers, then TLS is likely the better fit. If you need to build a large secure network, or create a secure tunnel between networks, then IPsec may make more sense.
If your use case combines aspects of both (a remote worker’s device connecting to a corporate VPN), then likely both TLS and IPsec will be needed. The reality is that modern networks are increasingly heterogeneous and distributed, and both IPsec and TLS will be needed for complementary layers of security in a Zero Trust posture. Thus IPsec and TLS should be considered synergistic rather than antagonistic security models.
If you have questions about any of the above, please contact us at facts@wolfSSL.com or +1 425 245 8247.
Download wolfSSL Now
