Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
Side-channel attacks recover secrets from the physical or timing behaviour of a system rather than from its algorithms, measuring how long an operation takes or how much power it draws.

What Are Side-Channel Attacks?
A cryptographic algorithm can be mathematically sound and still give its key away, because the implementation runs on physical hardware that behaves differently depending on the data it is processing. Side-channel attacks ignore the mathematics and measure the behaviour: how long an operation takes, how much current the chip draws, what it radiates, what it leaves in the cache. The secret is inferred from the measurement.
Newer entries keep the field active. Hertzbleed, disclosed in June 2022 as CVE-2022-23823 and CVE-2022-24436, turned dynamic frequency scaling into a remote timing channel, meaning a power side channel became observable over a network. GoFetch, published in March 2024, exploited the data memory-dependent prefetcher in Apple M-series processors to break constant-time cryptographic implementations that had previously been considered safe.
Threat model decides everything here. If an attacker can hold your device, as with a smart card, a payment terminal, a hardware wallet, or an embedded controller, physical side channels are a primary concern and the countermeasures belong in the product design. If your code runs on shared infrastructure alongside other tenants, microarchitectural channels matter, which is why cloud providers and browser vendors invest heavily in them. If you operate a conventional corporate estate, these attacks are far down the list behind phishing, unpatched services, and stolen credentials, though you still inherit the consequences through library and processor updates.
The core principle is that execution must not depend on secret data. In constant-time code, there are no branches on secret values, no table lookups indexed by secret values, and no early exits from comparisons. This is why cryptographic libraries are written the way they are, and why reimplementing a primitive in application code is a reliable way to introduce a side channel.
Beyond that, implementations use blinding, where a random factor is applied before the operation and removed afterwards so that the measured behaviour does not correlate with the key; masking, which splits secret values into random shares; and hardware countermeasures such as noise generation, dual-rail logic, and randomised clocking. Our note on side-channel resistant implementation covers these in detail.
For most organisations the practical actions are short. Use vetted cryptographic libraries and keep them updated, never write your own primitive, use crypto_memcmp-style constant-time comparison for secrets, apply microcode and firmware updates that address microarchitectural leaks, and for the highest-value key material use a hardware security module evaluated against power analysis. Where you are building a device rather than running one, our Hardware Penetration Test Services include power and timing analysis against the implementation you intend to ship.
No. Timing side channels appear in ordinary software whenever execution depends on secret data, such as a password comparison that stops at the first mismatching byte or a lookup table indexed by key bits. These are exploitable over a network in some cases and are fixed in code rather than in hardware, by writing comparisons and cryptographic operations to take the same time regardless of input.
Timing channels can, and some power channels now can too. Remote timing attacks against TLS implementations have been demonstrated repeatedly, and Hertzbleed showed that processor frequency scaling converts power consumption into an observable timing difference measurable over a network. Power, electromagnetic, and acoustic channels generally still require physical proximity to the target device.
It means the time taken and the memory accessed do not depend on secret values. In practice that rules out branching on a key bit, indexing a lookup table with secret data, and returning early from a comparison. Established cryptographic libraries are written this way and verified with tooling, which is the main reason implementing cryptographic primitives yourself is strongly discouraged.
Mainly through supply chain and patching rather than direct defence. Keep operating systems, microcode, and cryptographic libraries updated so that disclosed microarchitectural leaks are mitigated, and use evaluated hardware for key storage. Direct countermeasure work is for organisations building cryptographic hardware, embedded devices, or software where an attacker can run code on the same physical machine.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.