Session Reuse, Not a Cheaper Cipher, Makes TLS Affordable

Session Reuse, Not a Cheaper Cipher, Makes TLS Affordable

Ada re-derives this chapter’s own numbers step by step, at full precision

foundations
math-foundations
calculation-audit
mqtt
Ada ADA · CALCULATION AUDIT

Session Reuse, Not a Cheaper Cipher, Makes TLS Affordable

Turning on TLS adds a one-time handshake worth 0.72 mAs every time a sensor connects, while each message it sends costs about 0.4 mAs. Hold one connection open for a day of 1440 messages and that handshake is spread almost to nothing; reconnect for every reading and it is paid in full each time. This audit amortises the same 0.72 mAs both ways to test the claim in its title: is it session reuse, not a cheaper cipher, that makes TLS affordable?

Companion to the chapter MQTT Security Fundamentals — every number here comes from that chapter.

See the relationship before changing it

The figure reads from left to right. The blue card is messages per tls session. The middle card applies the page rule. The green card is handshake cost per message. Walk the arrows once: set the input, apply the rule, then read the result with its unit.

Messages per TLS session changes handshake cost per message An input card leads through the rule cost = 0.72 mAs / messages per session to the handshake cost per message result. INPUT PAGE INPUT APPLY THE RULE predict calculate check units OUTPUT RESULT
Walk the arrows. Session reuse spreads the same handshake charge across many readings.

Derive the baseline in four named moves

  1. 1

    Name the input. The chapter baseline is 1440 messages.

  2. 2

    Name the relationship. cost = 0.72 mAs / messages per session

  3. 3

    Substitute with units. 0.72 / 1,440 = 0.0005 mAs

  4. 4

    Read the result. Keep the unit beside the value. Use it only inside the technical boundary on this page.

Predict, then change messages per tls session

Try Predict the direction of cost = 0.72 mAs / messages per session. Test another messages per tls session, then compare handshake cost per message.

1440 messages
Chapter baseline
Handshake cost per message

Observe Session reuse spreads the same handshake charge across many readings. Reset messages per tls session to 1440 and compare handshake cost per message.

Explain Session reuse spreads the same handshake charge across many readings.

Check yourself

What should you do before trusting a moved-control result?
Answer: Predict its direction, apply the shown relationship, keep the units, and reset to the worked baseline.
What does this small model leave out?
Answer: Only messages per tls session moves here. Field effects named in the technical boundary stay fixed.
TryRecompute Ehandshake = 6 packets x (8 mA x 15 ms) = 6 x 120 mA-ms = 720 mA-ms = 0.72 mAs.
ObserveTrack Handshake share per message: 0.72 / 1440 = 0.0005 mAs.
ExplainExplain Per message: 0.72 + 0.4 = 1.12 mAs, of which the handshake is 0.72 / 1.12 = 64.3%.

Ready: use the stated baseline inputs, then compare each displayed result.

Ada: The callout and calculator both say the TLS handshake energy “falls away” over a session. Let me make that precise, because the same 0.72 mAs is either negligible or dominant depending entirely on how often you reconnect.

The handshake energy is a fixed, one-time cost per connection:

  • E_handshake = 6 packets x (8 mA x 15 ms) = 6 x 120 mA-ms = 720 mA-ms = 0.72 mAs

Amortise it over a one-day session of N = 1440 messages, each costing about 0.4 mAs to send:

  • Handshake share per message: 0.72 / 1440 = 0.0005 mAs
  • Total per message: 0.0005 + 0.4 = 0.4005 mAs
  • Handshake as a fraction of energy: 0.0005 / 0.4005 = 0.12%

Now reconnect for every single message instead, and each message pays the full handshake:

  • Per message: 0.72 + 0.4 = 1.12 mAs, of which the handshake is 0.72 / 1.12 = 64.3%

The cipher and the 0.72 mAs never changed; only the number of messages sharing one handshake did. Keeping the connection persistent drives the handshake from 64% of per-message energy down to 0.12%, so the design lever that makes TLS practical on a battery is session reuse, not a lighter TLS version, and the one-time 100 ms (TLS 1.3) or 200 ms (TLS 1.2) latency is paid once rather than on every reading.

Every number above is taken from the chapter’s own material and re-derived step by step.

Technical boundaries. This energy audit deliberately does not simulate cipher implementation, radio fading, certificate validation, or session-ticket expiry. It compares fixed handshake and resumed-session byte and energy costs from the chapter.