Why Do Computers Use So Much Energy?

Doing the math is cheap. Most of a computer's energy goes into moving data: from memory to the chip, from chip to chip, and then into the fans and chillers that carry the heat away.

Last updated September 2026
Figure 1 · The gap

Physics allows almost free computing. We are nowhere near it.

There is a theoretical minimum energy for erasing one bit of information, and it is astonishingly small. Real computers spend vastly more, and the biggest reason is distance: every time data travels from memory to a processor, it charges wires along the way.

kT ln 2The physical minimum heat to erase one bit, less than a billionth of a billionth of a joule at room temperature.
1.18 TB/sHow fast one modern high-bandwidth memory stack feeds a chip, per NIST. Memory speed is now a strategic resource.
$3.87bnOne company's planned investment in a single Indiana plant just to package this kind of memory.

Landauer's principle is a thermodynamic limit, not measured chip energy. Vendor bandwidth is a peak component figure, not application throughput or energy efficiency.

The story in one paragraph

For decades, computers got more efficient at a remarkable pace: the number of computations per kilowatt-hour doubled roughly every eighteen months. Much of that came from shrinking transistors, which made each switch cheaper. But arithmetic got cheap so fast that it stopped being the main cost. In a typical chip today, fetching a number from memory can take hundreds of times more energy than adding it to another number. So the frontier of efficient computing has moved from the transistor to the journey: how far each bit travels, how often, and how many are really needed. That is why memory stacked beside processors, optical links and smarter software now matter as much as faster chips, and why data-center electricity keeps climbing even as each operation gets cheaper.

  • The floor is remote: real systems run many orders of magnitude above the physical minimum.
  • Moving beats math: reading data from off-chip memory costs far more energy than computing with it.
  • Memory is a bottleneck: NIST now treats high-bandwidth memory as a core AI supply-chain constraint.
  • Demand still rises: the IEA expects global data-centre electricity to roughly double between 2025 and 2030.

Measured results, derived quantities, projections, targets, and editorial inference are identified by context. Announced capacity is never treated as operating performance.

Part I: Where the energy goes

A staircase from the physical floor to off-chip memory

Follow a single bit from storage to the arithmetic unit. Each step it takes costs more energy, and the biggest jumps happen long before it reaches a network cable.

Figure 2 · The floor-to-reality ladder

Every layer between compute and its data costs more than the Landauer floor

Every layer between compute and its data costs more than the Landauer floorEnergy per bit rises from about 2.85 zeptojoules at the Landauer floor to roughly 30 picojoules for off-chip DRAM access, a span of about ten orders of magnitude, plotted on a logarithmic scale across on-chip arithmetic, cache, HBM, DRAM, and network interconnect.1.0 zJ1000.0 zJ1 fJ1 pJ100 pJLandauer floor (300K)On-chip arithmeticOn-chip cache accessHBM package interfaceOff-chip DRAM accessNetwork interconnect123456Energy per bit, joules (log scale)
Landauer floor (300K) · 2.8 zJ

kT ln 2 at room temperature: the minimum thermodynamic dissipation for one irreversible bit erasure. No real device operates anywhere near this.

The Landauer point is a thermodynamic bound at 300 K. The arithmetic, cache, and DRAM points derive from Horowitz's historical 45 nm examples per 64-bit operation. HBM is a modeled HBM3 interface value; the optical point is a representative pluggable-link estimate. These are different measurement boundaries and generations, not one measured system. Sources: NIST: energy complexity of computation; Horowitz: Computing's Energy Problem (ISSCC 2014); University of California: HBM generation model; IEEE: optical interconnections.
Part II: The layers of a computer

Four layers, each deciding how far a bit must travel

Only the bottom layer is set by physics. Everything above it is design: where memory sits, how chips talk, and whether software reuses the data it already has.

01

Physical state

Transistors, memory cells and wires hold and flip bits while fighting electrical noise.

Measure
J/switch · error rate
Failure boundary
Leakage, reliability margins and the need for speed all cost energy far beyond the theoretical minimum.
Where the frontier moves

Devices that hold a bit reliably with less energy, without slowing down.

02

Package and memory

Caches, high-bandwidth memory stacks and the silicon they sit on trade storage capacity against distance from the processor.

Measure
J/bit · GB/s · latency
Failure boundary
Heat and manufacturing yield limit how close memory can sit to the processor.
Where the frontier moves

Advanced packaging that keeps shrinking the distance data travels.

03

System interconnect

Cables, switches and network links move data between chips, servers and racks.

Measure
Bytes/result · network utilization
Failure boundary
Once data leaves the package, network layout and waiting for other chips can dominate.
Where the frontier moves

Optical links built into the chip package, closing the gap with on-package wiring.

04

Algorithms and software

Number precision, reuse, caching and scheduling decide how much data needs to move at all.

Measure
Verified outcomes/joule
Failure boundary
Poorly organized software fetches or recomputes the same data again and again.
Where the frontier moves

Software that treats data movement, not arithmetic, as the thing to optimize.

Part III: Why the old trick stopped working

Erasing a bit has a floor. Moving one has a distance.

Until the mid-2000s, smaller transistors also used less power, so every new generation got efficiency for free. That automatic saving faded. Transistors kept shrinking, but wires did not get cheaper to charge, and real devices must hold their state firmly against noise. The practical gains now come from putting data closer and using it more, not from approaching the theoretical floor.

energy per bit moved × bytes+switching and facility energy=energy per verified result

Efficiency improved, and the workload grew faster

Koomey and colleagues’ long-run series found computations per kWh roughly doubled every 1.5 years. That is a historical trend for a particular benchmark and set of machines, not a law of nature, and it slowed once voltage scaling stalled after the mid-2000s.

Meanwhile, total demand took off. The IEA estimates 485 TWh of global data-centre electricity in 2025, about 1.5% of world demand, and roughly 950 TWh in its 2030 case. Those totals include ordinary servers, AI accelerators, storage, networks and cooling, so they cannot be divided by a guessed number of AI prompts to get a real energy-per-answer figure. This is the paradox of efficient computing: each operation gets cheaper, so we buy far more of them.

How far above the floor are we?

Landauer's minimum at room temperature is about 2.9 × 10⁻²¹ joule per bit erased. It applies to an ideal operation, not to a transistor, a memory read, a network link, a power converter and a cooling system combined. Researchers exploring “reversible” logic have shown that some of the gap can in principle be closed, but no one has built a general-purpose computer that does so. For the foreseeable future, the big wins are engineering ones: shorter wires, fewer bytes, and better reuse.

Part IV: How engineers are fighting back

The package becomes the computer

When memory sits right next to logic, the way chips are bonded, cooled, tested and supplied matters as much as the processor design. Four strategies are doing most of the work.

Reuse before moving

Reorganize algorithms so data fetched once is used many times while it is still nearby.

Use fewer bits

Store and move numbers at the lowest precision that still gives a correct answer.

Shorten the wire

Put memory right beside the processor with HBM, chiplets and optics where the workload justifies it.

Measure outcomes

Report energy per finished task for the whole system, not peak operations for one chip.

Who is building what

Co-packaged optics, silicon photonics, high-bandwidth memory and open chiplet standards all attack the same “memory wall.” Search the record, or filter by interconnect layer.

8 programmes
Celestial AIPhotonic FabricSilicon photonics platform delivering optical compute and memory interconnectivity, decoupling compute silicon from memory placement via optical links
Reported evidence
Secured major customer commitments and hyperscale investments; demonstrated multi-terabit low-latency die-to-die and die-to-memory interconnects.
Announced next step
Commercial deployment of disaggregated, optically pooled high-bandwidth memory (HBM) systems for multi-accelerator AI clusters.
Unresolved risk
Optical coupling losses at fiber-to-chip interfaces and manufacturing yield of dense silicon photonic waveguide arrays.
Ayar LabsTeraPHY & SuperNova LaserMonolithic electronic-photonic optical I/O chiplets utilizing micro-ring modulators driven by external multi-wavelength continuous-wave laser sources
Reported evidence
Demonstrated operational TeraPHY links integrated with Intel processors and NVIDIA GPUs; achieved <5 pJ/bit energy efficiency across multi-terabit links.
Announced next step
Integration of optical I/O directly into standard high-volume packaging substrates replacing electrical SerDes traces.
Unresolved risk
Thermal sensitivity of micro-ring modulators requiring active thermal tuning loops, and laser diode lifetime in data center environments.
LightmatterPassage Optical InterposerWafer-scale programmable silicon photonics interposer that routes optical signals between heterogeneous compute dies placed directly on top
Reported evidence
Demonstrated wafer-scale optical routing with orders-of-magnitude higher cross-sectional bandwidth density than organic packaging substrates.
Announced next step
Enabling rack-scale virtual compute chips consisting of hundreds of interconnected GPUs behaving as a single monolithic processor.
Unresolved risk
Mechanical stress and thermal expansion differences across full 300mm silicon interposers under high thermal dissipation loads.
BroadcomBailly 51.2 Tbps Co-Packaged OpticsCo-Packaged Optics (CPO) switch integrating 51.2 Tbps silicon switching ASICs with optical transceiver engines on a common substrate
Reported evidence
Manufactured and demonstrated working commercial 51.2 Tbps Bailly CPO systems, cutting optical interconnect power by over 30% versus pluggable transceivers.
Announced next step
Scaling CPO to 102.4 Tbps switch architectures, eliminating copper backplane trace losses in hyperscale AI network fabrics.
Unresolved risk
Serviceability and field repair: replacing a failed optical channel currently requires removing the entire multi-thousand-dollar switch package.
SK HynixHBM3E & HBM4 Memory Stacks12- and 16-high stacked DRAM dies interconnected via Through-Silicon Vias (TSVs) and Advanced Mass Reflow Molded Underfill (MR-MUF)
Reported evidence
Dominant supplier of HBM3E memory for NVIDIA H100 and B200 systems, delivering over 1.18 TB/s bandwidth per stack.
Announced next step
Transitioning to HBM4 featuring 2048-bit wide interfaces and logic base dies manufactured on advanced foundry nodes (TSMC).
Unresolved risk
Thermal dissipation from bottom DRAM layers trapped under top dies, and TSV micro-bump bonding yield across 16-layer stacks.
TSMC3DFabric & CoWoS (Co-Packaged Optics)Chip-on-Wafer-on-Substrate (CoWoS-R, CoWoS-L) and Compact Universal Photonic Engine (COUPE) integrating optics and compute in 3D packaging
Reported evidence
The manufacturing foundation for modern AI accelerators; expanding CoWoS capacity to multi-reticle dimensions exceeding 3.3× standard mask size.
Announced next step
Monolithic integration of electronic and photonic ICs (EPIC) with lowest parasitic capacitance and sub-picojoule transfer energy.
Unresolved risk
Substrate warpage during thermal cycling and severe global packaging substrate supply bottlenecks constraining AI accelerator shipments.
UCIe ConsortiumUniversal Chiplet Interconnect ExpressOpen industry standard defining physical, die-to-die adapter, and protocol layers (PCIe, CXL, streaming) for heterogeneously packaged chiplets
Reported evidence
Over 100 member companies including Intel, AMD, Arm, TSMC, and Qualcomm; validated interoperability across silicon vendor test chips.
Announced next step
Creating a thriving commercial merchant chiplet market where designers mix and match dies across different foundries and process nodes.
Unresolved risk
Testing and known-good-die (KGD) liability when chiplets from multiple competing vendors are integrated into a single expensive package.
IEEE Photonics / DARPA PIPESPhotonic Edge-to-Edge StandardsPublic research and standards initiatives benchmarking photonic link energy efficiency, latency, and optical interconnect architectures
Reported evidence
Funded foundational micro-ring, comb laser, and optical interposer developments that seeded leading commercial silicon photonics startups.
Announced next step
Establishing sub-1 pJ/bit interconnect metrics for exascale computing and disaggregated data center memory pools.
Unresolved risk
Academic prototypes demonstrating record efficiency in controlled settings struggling with manufacturing tolerances on commercial CMOS lines.

Optical I/O energy ratings (e.g. sub-5 pJ/bit) describe transceiver physical layer links; system-level energy must account for external continuous-wave laser thermal coolers and electrical serializer overhead.

The optimistic view, with conditions

The efficient computer moves less, not faster

The largest gains ahead come from designing algorithms, number formats, memory, packaging, networks and cooling together, so useful information travels less often and less far. There is enormous room left: we are still far closer to the cost of wires than to the limits of physics.

Package

Keep shortening the wire

High-bandwidth memory already beats ordinary memory per bit because it sits closer; advanced packaging pushes the same logic further.

Network

Bring light onto the chip

Optics built into the package can cut chip-to-chip energy well below today's plug-in modules.

Software

Stop paying twice for the same bit

Algorithms that keep data nearby avoid fetching it again and again.

What actually lowers the energy of a useful answer

  1. Shorter distancesMemory placed physically close to where data is used.
  2. Fewer bits movedPrecision and compression matched to what the task needs.
  3. More reuseAlgorithms and caches that avoid fetching the same data twice.
  4. Whole-system measurementEnergy and latency reported per finished task, not peak operations.
  5. Layers designed togetherAlgorithms, packaging and networks optimized as one system.

Sources, method, and boundaries

The thermodynamic floor follows NIST's discussion of conventional and reversible computation. HBM figures come from NIST's project summary. The energy-per-bit ladder combines this floor with Horowitz's widely cited 45nm energy breakdown (ISSCC 2014) and public HBM and data-center optical-interconnect figures: representative orders of magnitude spanning different technologies and process nodes, not measurements of one system. The roofline relationship is a model, not a benchmark for a specific application.

Landauer's principle
The thermodynamic minimum energy required to irreversibly erase one bit of information at a given temperature.
Roofline model
A performance model showing that achievable compute throughput is capped by memory bandwidth for a given arithmetic intensity.
Co-packaged optics
Optical transceivers integrated directly onto a chip package rather than connected via a separate pluggable module.

Read More

Essays, books, talks, and research that shaped this field’s arguments. Influence is not endorsement; company communications and advocacy are labeled. Some publisher links require a subscription.

  1. Talk

    Mark Horowitz — Computing’s Energy Problem (and what we can do about it)

    The widely reused energy table makes data movement visible beside arithmetic.

  2. Foundational paper

    Rolf Landauer — Irreversibility and Heat Generation in the Computing Process

    The thermodynamic bound; compare it with the much larger costs of deployed architectures.

  3. Company technical writing

    NVIDIA — Technical blog

    Architecture and memory-bandwidth accounts from a hardware supplier, labeled as company writing.

  4. Company technical writing

    Google — Research blog

    Hyperscaler accounts of computing systems and efficiency, to read alongside accepted-work measurements.

  5. Paper / report

    Wulf & McKee, “Hitting the Memory Wall: Implications of the Obvious” (1995)

    The classic essay/paper that made the processor–memory performance gap a central computing-system problem.

Across the fields: learning curves, deployment, and rebound