COMPUTE ATLAS Supercomputer supply-chain graph
1581 systems 558 sites Sourced data

Supercomputers/Analysis/Access

Analysis 18

Who can get time on a supercomputer

Of the 439 systems in service in this dataset that record how people get time on them, 37% take applications from outside researchers through an open call and another 19% can be rented by anyone who pays: 56% in all. Weighted by measured performance, across the 222 of them with an HPL result, the share an outsider could apply for or rent is 65%. Open calls are a public-sector route and rental a private-sector one, and the restricted remainder is concentrated in a few very large machines: 43 restricted systems hold 26% of that measured compute, led by El Capitan.

Most systems have no access record yet, including the largest. Only 573 of the 1581 systems in the dataset (36%) carry an access-route row, and 439 of the 1093 in service. Those rows cover 55% of the measured performance in service, so the rest is unknown, not zero. The largest measured system in service, LineShine, has no access record, and the gap is widest for AI labs and Industry operators, where 14% and 15% of systems are covered. A system also counts in the most open class its text supports, so a machine that gives only part of its time to a national call counts as open in full: the open share is an upper bound on the capacity outsiders actually receive.

Researched with AI assistance, checked by scripts. These rows were researched by AI agents reading public operator, laboratory and vendor sources, appended by scripts and checked by automated validation; no subject-matter expert has reviewed them (how this site is made). A mechanical second-reader pass has since fetched every cited source: 1,350 of 1,581 systems have a readable citation that names them, and 64 are genuinely weakly sourced. Every claim carries its source and a confidence tier. Treat anything below verified as a lead, not a citation.

  • Open allocation or rentable
  • Other routes (own users, restricted, lease, other)
Share of systems in service an outside researcher could apply for or rent, by operator sector Horizontal bars, one per operator sector, each split into the share of systems in service whose access record is an open allocation or a rental, and the share reached only by other routes. All sectors 439 systems All sectors: 55.6% of 439 systems in service could be applied for or rented All sectors: 55.6% of 439 systems in service could be applied for or rented 56% apply or rent Government labs 73 systems Government labs: 27.4% of 73 systems in service could be applied for or rented Government labs: 27.4% of 73 systems in service could be applied for or rented 27% apply or rent Research HPC 264 systems Research HPC: 59.1% of 264 systems in service could be applied for or rented Research HPC: 59.1% of 264 systems in service could be applied for or rented 59% apply or rent Big tech 39 systems Big tech: 74.4% of 39 systems in service could be applied for or rented Big tech: 74.4% of 39 systems in service could be applied for or rented 74% apply or rent Neoclouds 30 systems Neoclouds: 86.7% of 30 systems in service could be applied for or rented Neoclouds: 86.7% of 30 systems in service could be applied for or rented 87% apply or rent Industry 20 systems Industry: 40.0% of 20 systems in service could be applied for or rented Industry: 40.0% of 20 systems in service could be applied for or rented 40% apply or rent Life sciences 8 systems Life sciences: 37.5% of 8 systems in service could be applied for or rented Life sciences: 37.5% of 8 systems in service could be applied for or rented 38% apply or rent 0%25%50%75%100%
Each bar is the share of a sector's systems in service, among those with an access record, that take applications through an open call or can be rented. AI labs, Finance have fewer than five such systems and are left off the chart but kept in the table and the file.
Table view: Share of systems in service an outsider could apply for or rent
Operator sectorSystems in serviceOpen allocationRentableApply or rentMeasured systemsApply or rent, by Rmax
All sectors43936.9%18.7%55.6%22265.5%
Government labs7323.3%4.1%27.4%4731.8%
Research HPC26452.7%6.4%59.1%13489.5%
Big tech392.6%71.8%74.4%1597.0%
Neoclouds306.7%80.0%86.7%11100.0%
Industry205.0%35.0%40.0%110.0%
Life sciences825.0%12.5%37.5%2too few
How this was calculated

Each system's access-route text, written during research from public sources (460 of the 573 rows are operator-stated; the rest come from press reports, vendors or filings), is put into one class by an ordered list of 33 keyword rules, first match wins. Explicit single-tenant and no-outside-access statements are tested first, then named allocation programmes (ACCESS and its predecessors, INCITE, EuroHPC, PRACE, HPCI, NAIRR, AIRR and national services such as NAISS, Sigma2 or the Digital Research Alliance of Canada) and open-call wording, then paid and cloud access, then testbeds and early access, then own-institution wording, then internal and mission use. A state-wide scheme open to researchers at every university in a region counts as open. A university charging its own researchers counts as own users; one selling time to outside organisations counts as rentable. The last own-users rule is deliberately broad, which is why no system is left unclassified; its errors land in own users.

The rules were tuned by printing every one of the 573 raw texts with its class and reading them all. A sample of 40 systems drawn with a fixed seed was then read by hand and judged against the class definitions: 38 of 40 agreed. The rules were fixed for the 2 that did not, and a second sample of 20 different systems agreed in 20 of 20. The hardest boundaries are open allocation against own users for regional consortia, and rentable against open for national centres that both run calls and charge.

"In service" means a recorded status of operational. Performance weighting uses Rmax, the measured HPL result, for the systems that have one; a system without one counts in the system shares and not in the weighted shares. The class describes access as the record states it now, or when the system ran, and does not track a change of model over a system's life. The downloadable file carries counts and shares only.

The access classes

ClassWhat it meansSystems with a recordIn service Share of systems in serviceShare of measured Rmax in service
Open allocation to outside researchers Outside researchers can apply through a published call or allocation process (national or international programmes such as ACCESS, INCITE, EuroHPC, PRACE, HPCI, NAIRR or AIRR, or a national service open to every researcher in a country). 248 162 36.9% 49.0%
Commercial or paid rental Anyone who pays can use it: public cloud instances, GPU clouds, reserved capacity, or a centre that sells time to outside organisations. 94 82 18.7% 16.4%
Leased or contracted to one customer The whole system or campus is built for, leased to or contracted to a single customer. 11 9 2.1% none measured
Institutional: own users and collaborators The operating university, laboratory or consortium's own researchers, plus sponsored collaborators and members who buy in. 149 127 28.9% 6.5%
Restricted: classified, mission or company-internal Classified or defence work, operational mission use (weapons, weather, agency programmes) or a company running its own workloads, with no outside route described. 48 43 9.8% 26.3%
Other: testbeds, early access, mixed or not yet set Testbeds and early-access systems for selected teams, systems split between own use and a small number of named partners, and future systems whose access model is not yet described. 23 16 3.6% 1.8%

How much one machine decides

The weighted answer rests on few machines. Of the measured compute in service that has an access record, El Capitan alone holds 21% and the five largest systems 55%. Taking out each system in turn and recomputing, the largest swing comes from El Capitan: without it the share an outsider could apply for or rent rises from 65% to 82%, 17 points. The largest swing the other way is Aurora, whose removal lowers it to 61%, 4 points. So the system-count share is the steadier number, and the weighted one moves with a single machine: the restricted El Capitan is the main reason it is not higher.

The largest systems in each class

ClassLargest measured systems in serviceIn service without an Rmax
Open allocation to outside researchers Aurora (1.01 EFlop/s) ; JUPITER (1.00 EFlop/s) ; Fugaku (442.0 PFlop/s) ; Alps (434.9 PFlop/s) ; Isambard-AI Phase 2 (216.5 PFlop/s) 66
Commercial or paid rental Eagle (561.2 PFlop/s) ; CHIE-4 (135.4 PFlop/s) ; MAXIMUS-384 (114.5 PFlop/s) ; CHIE-2 (89.78 PFlop/s) ; Shakti Cloud (84.31 PFlop/s) 53
Leased or contracted to one customer none has a measured Rmax 9
Institutional: own users and collaborators Tuolumne (208.1 PFlop/s) ; Shaheen III - GPU (122.8 PFlop/s) ; Shaheen III (35.66 PFlop/s) ; HiPerGator AI (17.49 PFlop/s) ; Kempner AI Cluster (16.29 PFlop/s) 70
Restricted: classified, mission or company-internal El Capitan (1.81 EFlop/s) ; Eos (121.4 PFlop/s) ; Harpia (75.20 PFlop/s) ; El Dorado (68.02 PFlop/s) ; Crossroads (30.03 PFlop/s) 13
Other: testbeds, early access, mixed or not yet set Israel-1 (41.50 PFlop/s) ; Helma (32.22 PFlop/s) ; Polaris (25.81 PFlop/s) ; Taipei-1 (22.30 PFlop/s) ; Redtail (12.85 PFlop/s) 6

None of the 11 single-tenant systems with an access record has a measured Rmax: the AI campuses built for one customer have not run the benchmark, so the weighted shares on this page cannot see them at all.

  • Open allocation
  • Rentable
  • Single-tenant lease
  • Own users
  • Restricted
  • Other
Access class of systems in service, 2008–2026 Stacked area chart of the share of systems in service each year, among those with an access record, by how outsiders can get time on them. Open allocationRentableSingle-tenant leaseOwn usersRestrictedOther 0%25%50%75%100%2008201020122014201620182020202220242026 Open allocationRentableOwn usersRestrictedOther 2008 Open allocation 50.0% Rentable 3.8% Single-tenant lease 0.0% Own users 34.6% Restricted 7.7% Other 3.8%2009 Open allocation 51.9% Rentable 3.7% Single-tenant lease 0.0% Own users 33.3% Restricted 7.4% Other 3.7%2010 Open allocation 53.1% Rentable 3.1% Single-tenant lease 0.0% Own users 31.3% Restricted 9.4% Other 3.1%2011 Open allocation 54.1% Rentable 5.4% Single-tenant lease 0.0% Own users 27.0% Restricted 10.8% Other 2.7%2012 Open allocation 58.6% Rentable 3.4% Single-tenant lease 0.0% Own users 27.6% Restricted 6.9% Other 3.4%2013 Open allocation 60.0% Rentable 4.6% Single-tenant lease 0.0% Own users 26.2% Restricted 6.2% Other 3.1%2014 Open allocation 55.8% Rentable 3.9% Single-tenant lease 0.0% Own users 31.2% Restricted 6.5% Other 2.6%2015 Open allocation 56.3% Rentable 5.2% Single-tenant lease 0.0% Own users 29.2% Restricted 7.3% Other 2.1%2016 Open allocation 56.0% Rentable 4.6% Single-tenant lease 0.0% Own users 31.2% Restricted 6.4% Other 1.8%2017 Open allocation 54.5% Rentable 4.9% Single-tenant lease 0.0% Own users 31.7% Restricted 7.3% Other 1.6%2018 Open allocation 54.3% Rentable 5.8% Single-tenant lease 0.0% Own users 30.4% Restricted 8.0% Other 1.4%2019 Open allocation 52.3% Rentable 7.8% Single-tenant lease 0.0% Own users 30.7% Restricted 7.8% Other 1.3%2020 Open allocation 51.2% Rentable 8.3% Single-tenant lease 0.0% Own users 30.4% Restricted 8.3% Other 1.8%2021 Open allocation 52.8% Rentable 9.0% Single-tenant lease 0.5% Own users 28.1% Restricted 7.0% Other 2.5%2022 Open allocation 50.4% Rentable 11.2% Single-tenant lease 0.4% Own users 25.9% Restricted 8.5% Other 3.6%2023 Open allocation 49.2% Rentable 11.4% Single-tenant lease 0.4% Own users 24.4% Restricted 11.0% Other 3.5%2024 Open allocation 47.0% Rentable 13.8% Single-tenant lease 0.3% Own users 23.5% Restricted 11.4% Other 4.0%2025 Open allocation 43.6% Rentable 17.1% Single-tenant lease 1.4% Own users 22.8% Restricted 11.3% Other 3.8%2026 Open allocation 43.4% Rentable 18.1% Single-tenant lease 2.2% Own users 21.3% Restricted 11.1% Other 4.0%
The mix of access classes among systems in service each year, counting only systems with an access record and a first-operational date. It starts in 2008, the first year with at least 25 such systems in service.
Table view: Access class of systems in service, 2008–2026
YearOpen allocationRentableSingle-tenant leaseOwn usersRestrictedOtherSystems with a record
200850.0%3.8%0.0%34.6%7.7%3.8%26
200951.9%3.7%0.0%33.3%7.4%3.7%27
201053.1%3.1%0.0%31.3%9.4%3.1%32
201154.1%5.4%0.0%27.0%10.8%2.7%37
201258.6%3.4%0.0%27.6%6.9%3.4%58
201360.0%4.6%0.0%26.2%6.2%3.1%65
201455.8%3.9%0.0%31.2%6.5%2.6%77
201556.3%5.2%0.0%29.2%7.3%2.1%96
201656.0%4.6%0.0%31.2%6.4%1.8%109
201754.5%4.9%0.0%31.7%7.3%1.6%123
201854.3%5.8%0.0%30.4%8.0%1.4%138
201952.3%7.8%0.0%30.7%7.8%1.3%153
202051.2%8.3%0.0%30.4%8.3%1.8%168
202152.8%9.0%0.5%28.1%7.0%2.5%199
202250.4%11.2%0.4%25.9%8.5%3.6%224
202349.2%11.4%0.4%24.4%11.0%3.5%254
202447.0%13.8%0.3%23.5%11.4%4.0%298
202543.6%17.1%1.4%22.8%11.3%3.8%346
202643.4%18.1%2.2%21.3%11.1%4.0%371
How this was calculated

A system is in service in a year if it first ran in or before that year and was not decommissioned before it. Each system keeps the one class its access record gives it for every year, so the chart shows how the mix of machines changed, not how any one machine's access policy changed. Earlier years rest on fewer records, mostly long-lived research systems, so the shape is the part to read.

Among systems with a record, rentable machines were 5% of those in service in 2015 and 18% in 2026, while open allocation went from 56% to 43%. That is the arrival of GPU clouds and hyperscaler capacity among machines that were mostly public research systems, and it also reflects which systems have had their access researched so far.

By region

Region (systems in service)SystemsOpenRentable LeaseOwn usersRestrictedOther Apply or rentApply or rent, by Rmax
North America16021%16%5%33%21%5%37%42%
Europe15153%13%1%29%1%3%66%96%
Asia-Pacific9437%34%0%24%3%1%71%93%
Middle East1921%21%0%32%16%11%42%2%
South America1164%9%0%9%18%0%73%17%
Africa450%25%0%25%0%0%75%too few

By era of first operation

First light (all systems with a record)SystemsOpenRentable LeaseOwn usersRestrictedOther Apply or rentApply or rent, by Rmax
Before 20103352%3%0%36%6%3%55%87%
2010–20145761%4%0%28%5%2%65%77%
2015–20198452%11%0%29%8%0%63%86%
2020–20228751%15%1%18%8%7%66%94%
2023–20248039%20%0%18%19%5%59%53%
2025 onwards8435%32%8%13%8%4%67%87%
No first-light date14832%18%2%38%5%5%50%47%

Shares are of systems; the last column weights by Rmax and reads "too few" where a group has fewer than three measured systems. The era table includes retired systems and systems not yet running, which is why its totals differ from the in-service tables.

What the records do not cover

These are the largest measured systems in service with no access record. Until they are researched, the shares above describe the systems that have been, not the dataset.

SystemOperator sectorCountryRmax
LineShine Research HPC China 2.20 EFlop/s
HPC7 Industry Italy 571.5 PFlop/s
HPC6 Industry Italy 477.9 PFlop/s
ISEG2 Neoclouds Netherlands 202.4 PFlop/s
Discovery 6 Industry USA 164.2 PFlop/s
Maximus-02 Neoclouds USA 156.1 PFlop/s
SSC-24 Industry South Korea 106.2 PFlop/s
Venado Government labs USA 98.51 PFlop/s
Operator sectorSystemsWith an access recordCoverage
Government labs34810530%
Research HPC77334845%
Big tech1164841%
AI labs21314%
Neoclouds1383525%
Industry1592415%
Finance8225%
Life sciences18844%

29% of ranked systems and 47% of unranked ones have an access record. 173 of the systems with an access record also link to the operator's own access page.

The read

Who can get time depends mostly on who owns the machine. Among university and research-centre systems in service, 59% take outside applications or rent time, and 39% serve only their own researchers and collaborators. Government labs look different: 27% are reachable from outside, and 37% are restricted to weapons, defence, weather or agency work. In the private sector the route is payment, not peer review: 72% of big-tech systems and 80% of GPU-cloud systems with a record can be rented, and 13% and 13% respectively are built or leased for a single customer.

By measured performance the answer looks more open than by count, because the largest public machines are allocated by national calls. But the measured share leans on few systems, and it is blind to the AI campuses that do not run the benchmark and to the 45% of measured compute in service that has no access record yet. The honest summary is that, of the compute this dataset can measure and has an access record for, most is open to an outside researcher in principle, through a competitive call or a credit card; how much of it any one researcher can get in practice is a question about oversubscription that these records do not answer.