/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

Top500: the Intel-powered Aurora supercomputer, not yet fully operational, came second behind the AMD-powered Frontier, but did take first in an AI benchmark

Aurora faces stability issues from hardware failures, cooling malfunctions, and operational errors.

Tom's Hardware Paul Alcorn

Context & Ripple Effects

Aurora’s rise follows its installation by Argonne and Intel, after which it was expected to come online with major FP64 capacity; its latest ranking narrows the gap with the incumbent leader without displacing it. Aurora’s installation milestone made the system’s eventual production readiness a central test of Intel’s return to top-tier HPC.

Frontier had already retained the top Top500 position in the prior ranking, while AMD’s presence among leading systems had expanded. Frontier’s prior lead gives Aurora’s second-place result significance as a competitive milestone, but its operational issues qualify the result.

First-order effects

  • Aurora gives Intel a leading AI-benchmark result and a second-place Top500 system, while Frontier and AMD retain the top overall Top500 position.
  • Hardware failures, cooling problems, and operational errors limit Aurora’s immediate usefulness until the system can run reliably at full scale.

Second-order effects

  • The split result makes benchmark leadership less interchangeable: prospective HPC users must weigh AI performance against sustained operational availability rather than treating a single ranking as decisive.
  • Intel’s result strengthens the case for an alternative at the high end of HPC, while Frontier’s continued lead preserves AMD’s performance reference point for competing deployments.

Third-order effects

  • If this pattern persists, large-scale compute procurement will put greater weight on execution risk—hardware integration, cooling, and operations—alongside peak benchmark scores.
  • HPC competition is broadening from a single fastest-system contest toward differentiated leadership across traditional simulation performance, AI workloads, and deployable reliability.

The trend: Exascale computing is becoming a multi-dimensional contest in which AI throughput and operational maturity matter alongside headline Top500 performance.

Discussion

  • Phoronix Michael Larabel on x
    Intel-Powered Aurora Supercomputer Breaks The Exascale Barrier
  • @hpc_guru @hpc_guru on x
    63rd #Top500 is out! o #Frontier @ORNL remains #1 o #Aurora @argonne becomes the 2nd official #Exascale #supercomputer o Only new #Top10 system is #Alps @cscsch https://top500.org/... #HPC #AI #ISC24 [image]
  • @glennklockwood Glenn K. Lockwood on x
    Erich highlighting new entries in the Bottom490. The first MI300 GPUs are starting to appear, including the LLNL El Capitan early access system. The next Top500 will be an interesting one. Good for AMD on keeping NVIDIA on its toes. #ISC24 [image]
  • @mobile_dom Domenico Lamberti on x
    TLDR for the new Top500 list: -Frontier retains it's No.1 spot. -Aurora is about 200Pf RMax behind Frontier, using 16kW more power, 500,000 cores more than frontier, though it is about 270Pf higher Rpeak than frontier. Aurora is... not looking great [image]
  • @cscfi @cscfi on x
    ✨ LUMI supercomputer is again the fastest #supercomputer in Europe, and globally 5th fastest, according to the new Top500 list. Pan-European @LUMIhpc, located in CSC's data center in Kajaani, Finland, was ranked high on other lists as well. #hpc https://www.lumi-supercomputer.eu/…
  • @glennklockwood Glenn K. Lockwood on x
    Wonder why Microsoft isn't listed in performance share but integrators like Lenovo/Penguin are.
  • @bmabey Ben Mabey on x
    Our new supercomputer, BioHive-2, clocked in at #35 on the TOP500 list announced today. Check out the timelapse of what building a supercomputer in less than 3 weeks looks like. 😀
  • @simonmcs Simon McIntosh-Smith on x
    @HPC_Guru Delighted that the first part of Isambard-AI has debuted so high on the Green500 - #2! It also claimed #128 on the Top500, the first time an Isambard system has featured in either list. Great achievement by the whole Isambard team along with HPE, NVIDIA and Arm
  • @glennklockwood Glenn K. Lockwood on x
    Rick is saying a lot of the quiet part out loud here. Non-Top500 HPL scale runs were run at much smaller scale since ALCF's focus was on breaking the 1 EF barrier HPL-MxP was 90% of the system Intel has “bad” product names 😬 #ISC24 [image]
  • @hpc_guru @hpc_guru on x
    #Green500: all Top 3 machines are new to the list And all 3 are based on Nvidia #GH200 1. JEDI - 72.73 GF/w 2. Isambard-AI - 68.83 GF/w 3. Helios - 66.95 GF/w https://top500.org/... #HPC #AI #GPU #ISC24 [image]
  • @hpc_guru @hpc_guru on x
    #Top500 Vendor Share @Lenovo has the most systems @HPE has the most performance #HPC #AI #ISC24 [image]
  • @bristoluni @bristoluni on x
    Phase 1 of Isambard-AI, which will be the UK's most powerful supercomputer, has officially been switched on🎉 At top 2 of the ISC Green500 list, we're home to the most sustainable supercomputer in the UK and the 2nd most efficient in existence♻ @SciTechgovuk @HPE @NVIDIA [video]