JavaScript is not available.

Jun 22, 2022

I'm very excited for El Capitan and the deployment of the MI300 APU. An HPC APU has been part of AMD's vision for many years, dating back to concepts in AMD Research and the DOE's fast-forward programs. It has been years of effort and co-design to get us to this point.

Quote Tweet

HPC Guru

@HPC_Guru

Jun 21, 2022

Terry Quinn revealed that @AMD’s forthcoming #MI300 #APU would be the computational bedrock of El Capitan, an #Exascale #supercomputer which is slated for installation @Livermore_Lab in late 2023 hpcwire.com/2022/06/21/amd #HPC #AI #GPU via @TiffanyTrader

119

Introducing the 2023 BSSw Fellows

Thanks

for shining a spotlight!

Introducing the 2023 BSSw Fellows -

hpcwire.com

Jan. 13, 2023 — The 2023 class of BSSw Fellows will be recognized during the 2023 Exascale Computing Project Annual Meeting, January 17-20, 2023. The BSSw Fellowship Program gives recognition and...

Replying to

The Queen in the North!

GIF

read image description

ALT

Feb 2

The Supercomputer of the North!

Quote Tweet

LUMI supercomputer

@LUMIhpc

Feb 1

The 2nd pilot phase of LUMI is complete, system’s performance verified with accepted benchmarks, contract work completed and, therefore, LUMI is officially ready to serve European scientists in its full capacity! Read more from the blog: lumi-supercomputer.eu/lumi-is-offici #eurohpc #hpc

0:53

627 views

Sarom Leang

@ifndef_define

Jan 17

Looking forward to seeing everyone in person!

Sunita Chandrasekaran

@sunitachandra29

Jan 14

Thank you Gina for sharing! This has certainly been a tremendous experience for us! Looking forward to further exploring #Frontier :-) #PIConGPU

#CAAR

Quote Tweet

A team studying plasma physics used #exascale supercomputer Frontier to achieve an eightfold speed increase in their code’s performance. The result is an exciting demonstration of what we can expect from Frontier in the years to come. @ORNL @OLCFGOV olcf.ornl.gov/2022/11/18/exa

Hylleraas Centre for Quantum Molecular Sciences and LUMI supercomputer

Sigbjørn Løland Bore

@SigbjornBore

Jan 12

So, it begins.....

Jan 11

What makes #exascale unique? Fun opportunity to articulate the many architectural advances in

@ORNL

Frontier. A system co-designed w/

@doescience

and

@HPE_HPC

for cutting edge computational science and ML/AI. It's been the project of a lifetime!

Quote Tweet

Right before #SC22 I had a great conversation with @nicholasmalaya on Scaling with @AMDInstinct and Frontier. Summarized conversation & thoughts into a new blog: “AMD Instinct: Scaling the Heights of #GPU Acceleration for a New Supercomputing Era” bit.ly/Instinct_Scali

Satoshi Matsuoka

Myths and Legends in High-Performance Computing

Jan 9

The 'Myths' paper is out: Satoshi Matsuoka, Jens Domke, Mohamed Wahib, Aleksandr Drozd, Torsten Hoefler. "Myths and Legends in High-Performance Computing" arxiv.org/abs/2301.02432" undergoing peer review. Hopefully the community will be engaged in yes/no discussions on Twitter!

arxiv.org

In this humorous and thought provoking article, we discuss certain myths and legends that are folklore among members of the high-performance computing community. We collected those myths from...

137

366

Lawrence Livermore National Laboratory

@Livermore_Lab

Jan 8

The next generation of #computing is here, in the form of a machine called 'El Capitan' and we're taking things to the next level in 2023. But how will El Capitan change the landscape of science and technology at #LLNL? Stay tuned... youtu.be/fNLc0Vo-4tw

@Livermore_Comp

youtube.com

LLNL will scale new heights with El Capitan

As home to one of the world’s premier high performance computing facilities, we’re taking things to whole new level in 2023. Exascale! With advanced processo...

Underfox

@Underfox3

Jan 5

After so many years of development and expectation... The EHP Era has arrived! 🦊🔥 Fig. 1 - My small sketch consistent with AMD patents from 2021 Fig. 2 - MI300 Exascale APU unveiled today

Replying to

Respectfully, Grace+Hopper is two chips with a fast custom interconnect and a software layer to present them in a more unified fashion. MI300 is unified at the hardware level. We at AMD believe the APU is a paradigm shift over previous tightly integrated devices.

Jan 5

...and HPC! A beast for computational discovery.

Quote Tweet

Sasa Marinkovic

@SasaMarinkovic

Jan 5

Step function leap on AI @AMD

Finite Difference Method - Laplacian Part 2 (amd-lab-notes)

AMD GPUOpen

@GPUOpen

Jan 4

#AMDlabnotes kicks off the new year🎆with the second part of our blog series on the finite-difference method for the Laplace operator. Here we present useful optimizations like loop tiling and reordering access patterns. #HPC #HIP #ROCm

gpuopen.com

In this post we introduce two common optimizations that can be applied to the kernel to reduce data movement and bring us closer to the new peak: loop tiling to explicitly reduce memory loads and...

Jan 1

Me: why would anyone stay up until midnight on new year's eve? Also Me: I'm not sleeping until the

@thoefler

paper drops on arXiv

Quote Tweet

Satoshi Matsuoka

Dec 31, 2022

The IEEE Fernbach & Cray 2022 awardees @thoefler & I, along w/the Riken R-CCS skunkworks aces, teamed up to write up what could become one of the most controversial paper on HPC in recent times. We just submitted it & posting the first draft on arXiv in a few days. Stay tuned!

Martin Bauer

@martinmbauer

Dec 25, 2022

It pains me to see eminent colleagues like Neil deGrasse Tyson propagate elementary mistakes. What would really be the implications of taking Santa Claus seriously? I’ll repeat the basics of Santa Mechanics. A theory way more advanced than Neil let’s on here. 🎄🧵 1/13

Quote Tweet

Neil deGrasse Tyson

@neiltyson

Dec 24, 2022

In case you were wondering, for Santa, in a single 24-hr Earth-night, to deliver presents to all those who celebrate Christmas, he must visit 25,000 homes per second.

1,539

8,085

Great to see increased recognition for this vital part of computational science! The community must recognize that writing scientific software *is* part of the scientific process. Cite the software you use in a paper! Contributing to OSS should help w/ promotions/tenure reviews.

Quote Tweet

Dr. Hai Ah Nam

@DrHaiAhNam

Dec 21, 2022

- If we want quality scientific software, it is vital that we give credit and kudos to those who make it happen every day. Glad @exascaleproject @NSF make this fellowship possible to shine a spotlight on our awesome colleagues. twitter.com/BetterSSw/stat…

Dec 17, 2022

This offends me too!

Quote Tweet

hardmaru

@hardmaru

Dec 17, 2022

“It is the more personal pain of seeing a 5% GPU utilization number in production. I am offended by it.”

—@ID_AA_Carmack, reflecting on his resignation from Facebook.

hardmaru

@hardmaru

Dec 17, 2022

“It is the more personal pain of seeing a 5% GPU utilization number in production. I am offended by it.” 🔥 —

@ID_AA_Carmack

, reflecting on his resignation from Facebook.

549

Filippo Spiga

@filippospiga

Dec 16, 2022

You learn the secrets of your Craft by shadowing the #Masters

Brian Spears

@bkspears9

Dec 13, 2022

So now we can share the details!! We predicted predicted this #fusion ignition event in advance. We used a combination of HPC simulation, Bayesian inference, #AI, and transfer learning to call it. We found that the most likely outcome would be ignition. It happened!!

453

U.S. Department of Energy

@ENERGY

Dec 13, 2022

BREAKING NEWS: This is an announcement that has been decades in the making. On December 5, 2022 a team from DOE's

@Livermore_Lab

made history by achieving fusion ignition. This breakthrough will change the future of clean power and America’s national defense forever.

1:28

2.8M views

From

Lawrence Livermore National Laboratory

2,323

26.6K

79.9K

U.S. Department of Energy

📣 TOMORROW:

, alongside

, will announced a major scientific breakthrough accomplished by researchers at our

@Livermore_Lab

. You don't want to miss this! Tune in LIVE at 10:00 AM ET: energy.gov/live

TOMORROW: Secretary Jennifer Granholm, alongside NNSA Administrator Jill Hruby, will announced a major scientific breakthrough accomplished by researchers at our Lawrence Livermore National Lab.

read image description

ALT

NNSA Administrator and 5 others

343

864

Replying to

Glad to see Microsoft showcasing an all AMD (CPU+GPU) server in their booth. Not too many people picked up on this and glad Microsoft broke out this 8 GPU MI250 Instinct monster to display to attendees. #SC22

glennklockwood.blogspot.com

Glenn K. Lockwood

@glennklockwood

Nov 24, 2022

I collected some of my notes and thoughts from SC22 last week and posted them here: glennklockwood.blogspot.com/2022/11/sc22-r Happy Thanksgiving to those of you celebrating. For my colleagues abroad, have a great week!

SC'22 Recap

The biggest annual conference in HPC, the SC conference , was recently held in Dallas, Texas in its second hybrid incarnation since being al...

Andrew "The Metaverse Guy" Steinwold

HPC Guru

@HPC_Guru

Nov 22, 2022

Fire your nuclear reactors! If claim of 100 trillion parameters is true, the compute power (even with specialized hardware) needed to train these models will be humongous Lots of vendor marketing folks might be salivating thinking of this 😜 #HPC #AI via

@NicoleHemsoth

Quote Tweet

@AndrewSteinwold

Nov 22, 2022

GPT-4 is rumored to be coming soon, sometime between Dec - Feb - GPT-3 has 175 billion parameters - GPT-4 supposedly has 100 trillion parameters It is something like 500x more powerful than GPT-3 What kinda stuff will you be able to create with GPT-4!?

Show this thread

Scientists Using Frontier Supercomputer Win 2022 Gordon Bell Prize, Another Frontier Team Named...

Nov 18, 2022

Gordon Bell Prize Winning team: "Clang compiler in the public release of ROCm did its job. We hope it leaves no doubt about the strength of AMD’s commitment to open-source software." Still work to do on software, but we have real science on AMD ROCm!

insidehpc.com

[SPONSORED CONTENT] How many researchers can say they’ve not only run their scientific job on the AMD-powered Frontier supercomputer, the world’s no. 1 [...]

Sunita Chandrasekaran

@sunitachandra29

Nov 18, 2022

While twitter is still there 🥁🥁🥁 #PIConGPU runs on FRONTIER 😎 tinyurl.com/2zyt7d6c We met the FOM #CAAR target 🍾 SP case study 4.44x & DP ~8x to summit

TY 🥳

et. al

Verónica Vergara (she/her/ella)

@verolero86

Nov 18, 2022

Last day of

@Supercomputing

#SC22! Come join us for the 9th workshop on accelerator programming using directives (WACCPD)! We have an exciting program. See you in D174! #HPC

The awards were so numerous they had to let some scientists into the photo to hold them all!😉

Quote Tweet

Congratulations to @AMD for winning nine #HPCwireRCA22 awards! hpcwire.com/2022-hpcwire-a

Great result, congratulations to the team. It was an impressive result and the team ran on #AMD CPU and GPU hardware. The student cluster competition is one of the best parts of #SC22

Quote Tweet

SCC Linpack winner #SC22

Show this thread

Satoshi Matsuoka

A joint accolade for #Frontier and #Fugaku & other top machines for this year’s Gordon Bell Prize! For #Fugaku this is two years in a row following our COVID-19 aerosol work. In fact this is the first time a non-JP team has won the GB prize using a flagship JP supercomputer.

Quote Tweet

HPCwire

@HPCwire

SC22: Gordon Bell Prize Goes to Plasma Accelerator Research ow.ly/quZh50LHoWU #HPC #SC22

Berkeley Lab Computing Sciences Area

Quote Tweet

@LBNLcs