All corrections
Wikipedia May 9, 2026 at 05:43 AM

en.wikipedia.org/wiki/Aurora_(supercomputer)

3 corrections found

1
Claim
It was briefly the second fastest supercomputer in the world from November 2023 to June 2024.
Correction

Aurora remained No. 2 on the TOP500 through the June 2024 list and did not drop from second place until the November 2024 list.

Full reasoning

This date range is too short. The official TOP500 list for June 2024 still ranks Aurora No. 2. Aurora only moved down later, when the November 2024 TOP500 list ranked El Capitan No. 1, Frontier No. 2, and Aurora No. 3. So Aurora was the second-fastest system from November 2023 until November 2024, not only until June 2024.

2 sources
2
Claim
In May 2024, Aurora appeared at number two on the Top500 supercomputer list
Correction

Aurora reached No. 2 on the June 2024 TOP500 list, not in May 2024.

Full reasoning

The official TOP500 materials identify this ranking as part of the June 2024 list, not a May 2024 list. TOP500’s own lists index says the 63rd TOP500 List was published in June 2024, and the June 2024 page says Aurora claimed the No. 2 spot with 1.012 EFlop/s. That makes the month in this sentence incorrect.

2 sources
  • TOP500 Lists Index | TOP500

    June 2024 The 63rd TOP500 List was published June 1, 2024 in Hamburg, Germany.

  • June 2024 | TOP500

    Also like the last list, the Aurora system at the Argonne Leadership Computing Facility in Illinois, USA, has claimed the No. 2 spot on the TOP500 ... with an HPL score of 1.012 EFlop/s.

3
Claim
It has around 10 petabytes of memory and 230 petabytes of storage.
Correction

Aurora’s storage is 230 PB, but its total memory is about 20.4 PB, not 'around 10 PB.'

Full reasoning

The storage figure is consistent with Argonne sources, but the memory figure is not. Argonne’s Aurora page lists Aggregate System Memory: 20.4 PB. Argonne’s Aurora user guide also breaks that memory down as 10.375 PiB CPU DRAM + 1.328 PiB CPU HBM + 7.968 PiB GPU HBM, which totals roughly 19.7 PiB (about 21 PB in decimal units). Either way, the machine has roughly 20 PB of memory, not around 10 PB.

2 sources
Model: OPENAI_GPT_5 Prompt: v1.16.0