惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Security Archives - TechRepublic
Security Archives - TechRepublic
I
InfoQ
阮一峰的网络日志
阮一峰的网络日志
云风的 BLOG
云风的 BLOG
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
AWS News Blog
AWS News Blog
S
SegmentFault 最新的问题
T
Tailwind CSS Blog
The Hacker News
The Hacker News
GbyAI
GbyAI
P
Palo Alto Networks Blog
博客园 - 三生石上(FineUI控件)
Y
Y Combinator Blog
Stack Overflow Blog
Stack Overflow Blog
博客园 - Franky
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
Cyberwarzone
Cyberwarzone
H
Help Net Security
S
Securelist
月光博客
月光博客
博客园 - 【当耐特】
T
Threatpost
T
Tenable Blog
G
GRAHAM CLULEY
博客园 - 司徒正美
I
Intezer
MyScale Blog
MyScale Blog
T
Threat Research - Cisco Blogs
P
Privacy & Cybersecurity Law Blog
The GitHub Blog
The GitHub Blog
C
CERT Recently Published Vulnerability Notes
T
Tor Project blog
Google DeepMind News
Google DeepMind News
C
Cybersecurity and Infrastructure Security Agency CISA
罗磊的独立博客
腾讯CDC
P
Privacy International News Feed
博客园_首页
The Cloudflare Blog
Cisco Talos Blog
Cisco Talos Blog
A
About on SuperTechFans
V
Vulnerabilities – Threatpost
A
Arctic Wolf
B
Blog RSS Feed
Recorded Future
Recorded Future
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
Google DeepMind News
Google DeepMind News
S
Security Affairs
Microsoft Security Blog
Microsoft Security Blog
L
LangChain Blog

The Register - Special Features: Supercomputing Month

GPUs aren't worth their weight in gold HPC won't be x86 forever – and it's starting to show The exascale offensive: America's race to rule AI HPC India satisfies its supercomputing needs, not its ambitions UK lines up £250M cloud procurement for AI research How supercomputer filesystem DAOS breaks out of its niche Eviden to build France's first exascale rig with AMD chips Power: The answer to and source of all your DC dilemmas GPU monsters eat supercomputing, legacy storage starves HPE details Vera Rubin blades for next-gen Cray Battery trade war hits booming datacenter market AI isn't throttling HPC. It is HPC Oak Ridge lab gets $125M to combine HPCs with quantum Power crunch threatens to derail AI datacenter construction $10B + spent on liquid cooling this week – it's only Tuesday Nvidia, OpenAI, and the trillion-dollar loop Nvidia will help build 7 AI supercomputers for for DoE HPE to build Discovery exascale successor for Oak Ridge NextSilicon eyes HPC market with Maverick-2 accelerators UK waves £750M supercomputer contract at HPC builders Tsunami forecasting to get faster thanks to El Capitan
Norway's new supercomputer to use waste heat to raise salmon
2025-11-27 · via The Register - Special Features: Supercomputing Month

Supercomputing Month

Norway's most powerful supercomputer will use waste heat to raise salmon

HPE-built system mixes Nvidia's Grace-Hopper superchips with AMD Turin CPUs to maximize HPC potential

This week the Norwegian scientific community celebrated the completion of the Olivia supercomputer, which combines AMD CPUs with Nvidia Superchips to deliver a 16-fold boost to the nation's computing capacity – and eventually put fresh fish on the table.

Built by Hewlett Packard Enterprise (HPE) for Norway's national scientific computing division Sigma2, and located in an underground datacenter deep within the Lefdal Mines, Olivia combines 504 of AMD's latest Turin CPUs and 304 of Nvidia's Grace Hopper Superchips with 5.3 petabytes of HPE Lustre storage via HPE's 200 Gbps Slingshot 11 NICs

Nvidia's Grace Hopper may not be its latest accelerators, but they have proven to be among the most energy-efficient ever made. The parts combine Nvidia's 72-core Grace CPUs with a 144 GB H100 graphics accelerator. Each of these 1,000-watt chips is capable of delivering upwards of 67 teraFLOPS of FP64 matrix math for highly precise scientific workloads, or as much as 4 petaFLOPS of sparse FP8 for machine learning tasks.

REG AD

In this fall's Top500 ranking of the most powerful publicly known supers, Olivia's GPU partition took 134th place, delivering 13.2 petaFLOPS in the High-Performance Linpack (HPL) benchmark.

REG AD

The combination of the Nvidia-based GPU partition and the AMD-based CPU section allows the machine to address a broader range of high-performance computing and AI workloads. While many scientific workloads benefit heavily from the highly parallel processing afforded by GPUs, not all do. The two partitions enable Olivia to address both possibilities.

Olivia is expected to support research into areas like renewable energy including hydro electric power, climate modeling, marine science, health, and language. Access to the machine won't be limited to the boffins at Sigma2 either, but will include researchers across the nation.

On top of being Norway's most powerful supercomputer to date, the system is also among its most sustainable, cutting power consumption by 30 percent compared to its predecessor Betzy. Eventually, Sigma2 aims to reuse waste heat generated by the liquid cooled Cray EX4000 compute blades to warm water for local salmon farms.

"Olivia marks a new chapter for Norwegian research. By combining cutting-edge performance with the unique energy efficiency of the Lefdal Mine Datacenter, we're not only accelerating scientific discovery, we're doing so responsibly," Helge Stranden, senior advisor for HPC and storage at Sigma2 said in a statement. ®