- HPC and Advanced Computing 2026
- Chapter 3
Systems and sectors at baseline
Contents
- ·Edition home
- ·Executive summary
- 1HPC is power, cooling and policy before it is speed
- 2Key findings
- 3Systems and sectors at baseline
- 4Risks and capability
- 5Power and siting limits meet fast-growing demand
- 6Readings from the other desks
- 7The 2027 agenda
- 8Systems in depth
- AMethodology
- BData appendix
- CReferences and record
- ·Complete edition, for print
3.1Where each system stood at evidence close
INF · Systems
LineShine Linpack result, new first place
2.198exaflopsCPU-only; about 42.2 MW.
Reported2026TOP5004
INF · Systems
Combined performance of the TOP500 list
18.73exaflopsup from 14.99 six months before.
Reported2026TOP5005
ENE · Efficiency
Green500 leader, KAIROS
73.282GFlops per watt3.05 petaflops at 46 kW.
ENE · Energy
US data centre electricity use
176TWh4.4% of US electricity.
Reported2023DOE12
FIN · Finance
Exascale Computing Project budget
1.8USD billionfrom 2016 to 2024.
Reported2024ECP17
GOV · Policy
EuroHPC budget, 2021 to 2027
8.2EUR billionat least; page gives the amended figure.
Reported2026EuroHPC11
3.2Domain posture
Each of the twelve coupled domains is read through HPC, in words, not scored. The coupling column names the dependency that carries HPC risk into each domain.
| Domain | Posture at baseline | Coupling | Status |
|---|---|---|---|
| WAT | Cooling-water use of large facilities is not bound in this edition. | ENE → WAT | Expert input EQL 1 |
| ENE | The five largest systems draw 15.8 to 42.2 megawatts; US data centres used 4.4% of electricity in 2023. | ENE ↔ INF | Reported EQL 4 |
| FOO | No documented coupling at baseline; weather and crop applications are not evidenced here. | none documented | Expert input EQL 1 |
| HLT | No documented coupling at baseline; drug discovery and genomics workloads are not evidenced here. | none documented | Expert input EQL 1 |
| BIO | No documented coupling at baseline. | none documented | Expert input EQL 1 |
| CLI | Electricity and cooling demand matter for emissions; the footprint of large systems is not bound. | INF → CLI | Expert input EQL 1 |
| INF | The core system: five systems at or above one exaflop; list performance 18.73 exaflops. | the core system | Reported EQL 4 |
| FIN | USD 1.8 billion for the US exascale project; USD 50 billion for CHIPS programmes; EUR 8.2 billion for EuroHPC. | FIN → INF | Reported EQL 4 |
| DAT | Large systems hold sensitive scientific data; no figure is bound. | INF ↔ DAT | Expert input EQL 1 |
| AIC | Accelerators are in 276 of 500 listed systems; the AI and scientific demand split is not bound. | INF ↔ AIC | Reported EQL 3 |
| MIN | No supply figure for chips, memory or minerals is bound for HPC. | MIN → INF | Expert input EQL 1 |
| TRD | US licence rules and guidance shape access to advanced computing items. | INF ↔ TRD | Reported EQL 3 |
3.3Sector screens
All sixteen sectors are screened for their dependence on HPC, on the same grammar as every national edition. The evidence column describes coverage of the baseline evidence, not the performance of the sector.
| Code | Sector | Baseline evidence | Exposure to watch | Evidence coverage |
|---|---|---|---|---|
| GS-00 | Cross-system | HPC couples to energy, chips and policy | power and siting limits | THIN |
| GS-01 | Water | cooling-water use not bound | facility siting | THIN |
| GS-02 | Energy | five largest systems draw 15.8 to 42.2 MW1 | load from large facilities | PARTIAL |
| GS-03 | Food | no documented coupling | none documented | THIN |
| GS-04 | Health | no documented coupling | none documented | THIN |
| GS-05 | Finance | public programmes of USD 1.8 billion and EUR 8.2 billion1711 | dependence on a few programmes | PARTIAL |
| GS-06 | Development | access to compute and skills | widening gaps | THIN |
| GS-07 | Climate | footprint at scale not bound | electricity demand | THIN |
| GS-08 | Disaster | no documented coupling | none documented | THIN |
| GS-09 | Infrastructure | warm-water cooling and heat reuse at JUPITER9 | single-site dependence | THIN |
| GS-10 | Cities | heat reuse into a campus network9 | none documented | THIN |
| GS-11 | Biodiversity | no documented coupling | none documented | THIN |
| GS-12 | Earth observation | weather and climate workloads not evidenced here | none documented | THIN |
| GS-13 | Cyber and digital | security of large systems not bound | access control and export rules | THIN |
| GS-14 | Policy | BIS notices, EuroHPC and UK plan131123 | fragmented rules | PARTIAL |
| GS-15 | Science | public lists and standards forums11516 | common benchmarks | GOOD |
Notes
Numbers are the edition’s reference numbers, so a note reads the same in every chapter. The full list, with every source and its licence, is in Annex C. Annex C, references
- 4TOP500 (ed.), LineShine Debuts at No. 1 as the TOP500 Enters a New Global Exascale Era (news release, 23 June 2026; LineShine, HPCG, HPL-MxP, Green500 values). TOP500 project, Hamburg. Observation year 2026. Accessed 2026-11-03.back
- 5TOP500 (ed.), Highlights, June 2026 (list totals, accelerator and processor shares, HPCG and HPL-MxP tables; the page gives some figures that differ from the list page). TOP500 project, Hamburg. Observation year 2026. Accessed 2026-11-03.back
- 3TOP500 (ed.), Green500 list, June 2026 (list page; top five by gigaflops per watt). TOP500 project, Hamburg. Observation year 2026. Accessed 2026-11-03.back
- 12US Department of Energy (ed.), DOE releases new report evaluating increase in electricity demand from data centres (news item, 20 December 2024; 176 TWh in 2023, 4.4% of US electricity; 6.7% to 12% projected for 2028). US Department of Energy, Washington. Observation year 2023. Accessed 2026-11-03.back
- 17Exascale Computing Project (ed.), project site (2016 to 2024; USD 1.8 billion; about 2,800 collaborators; concluded). US Department of Energy, Oak Ridge. Observation year 2024. Accessed 2026-11-03.back
- 11EuroHPC Joint Undertaking (ed.), Discover EuroHPC JU (about page; budget of at least EUR 8.2 billion for 2021 to 2027; Council Regulations 2021/1173, 2024/1732 and 2026/150). EuroHPC JU, Luxembourg. Observation year 2026. Accessed 2026-11-03.back
- 1TOP500 (ed.), TOP500 list, June 2026, 67th edition (list page; top systems, Rmax, Rpeak, cores and power). TOP500 project, Hamburg. Observation year 2026. Accessed 2026-11-03.back
- 9EuroHPC Joint Undertaking (ed.), JUPITER launching Europe’s exascale era (news item, 5 September 2025; about 24,000 GH200 superchips; warm-water cooling; funding shares; power and budget not stated). EuroHPC JU, Luxembourg. Observation year 2025. Accessed 2026-11-03.back
- 13US Department of Commerce Bureau of Industry and Security (ed.), BIS homepage (notice of 16 January 2025 on advanced computing integrated circuits; extension of 7 April 2026 to 31 December 2026; guidance of May 2026 on Country Group D:5 and Macau; full texts not opened). BIS, Washington. Observation year 2026. Accessed 2026-11-03.back
- 23UK Government (ed.), AI Opportunities Action Plan (page; expand the AI Research Resource by at least 20 times by 2030; non-AI HPC and exascale flagged as unresolved; no budget figure). Department for Science, Innovation and Technology, London. Observation year 2025. Accessed 2026-11-03.back
- 15OpenMP Architecture Review Board (ed.), OpenMP specifications (index page; version 6.0 of November 2024; technical report 15 as a public comment draft of 6.1, July 2026). OpenMP ARB, Urbana. Observation year 2026. Accessed 2026-11-03.back
- 16MPI Forum (ed.), MPI documents (index page; MPI 5.0 voted through 5 June 2025; MPI 4.1 of 2 November 2023). Message Passing Interface Forum. Observation year 2025. Accessed 2026-11-03.back
