liveBook · Manning

Cache attack

As explained in Real-World Cryptography

A cache attack infers a victim's secret-dependent computation by observing the timing or state changes of shared processor memory caches.

A cache attack turns shared processor memory into a spy. A malicious program watches how quickly selected code or data can be reread, revealing which parts a victim’s cryptographic computation recently used. section 13.4

A typical attack proceeds in three steps:

  • Share a machine with the victim, such as a cloud host running multiple virtual servers. section 13.4
  • Evict selected portions of a shared cryptographic library from the cache. section 13.4
  • Measure rereading time: a slow access suggests the victim did not use the portion, while a fast access suggests it repopulated the cache. section 13.4

Shared cryptographic service

A malicious virtual server monitors cache lines used by an OpenSSL-based service. Repeated timing measurements produce a computation trace resembling a power trace, which can expose secret-dependent behavior. section 13.4

How Real-World Cryptography covers it

How deep it goes
Explains cache hierarchy, eviction, rereading, and timing interpretation, but defers detailed attack resources and implementation defenses.
Examples
Uses cloud co-tenancy, shared OpenSSL libraries, cache eviction, and repeated rereading measurements as concrete examples.
What it assumes
Assumes basic familiarity with computer memory and cryptographic software while explaining cache layers from their timing behavior.
What you can do after
Evaluate how shared-cache timing can reveal a victim’s secret-dependent execution trace.
How it connects
Links cache attacks to side-channel attacks, power analysis, cryptographic implementations, and leakage-resilient cryptography.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Real-World Cryptography. The section links point to the parts of the book it draws on.