SPRT

252 posts

SPRT banner
SPRT

SPRT

@SPRT_RCCS

Hiring postdocs and/or researchers for Supercomputing Performance Research Team @ RIKEN R-CCS ( 理研 高性能計算モデリング研究チーム )

Katılım Temmuz 2022
75 Takip Edilen48 Takipçiler
Sabitlenmiş Tweet
SPRT
SPRT@SPRT_RCCS·
Hi @IanCutress , @intelhpc here are a few more details regarding the myth to see zetta any time soon. We would love to be wrong, tho.
Satoshi Matsuoka@ProfMatsuoka

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!

English
1
0
3
2.8K
SPRT
SPRT@SPRT_RCCS·
For a few countries, this salary might not look like much, but in Japan it is! Ignore current exchange rates (which are out of whack), but look at cost of living, convenience, and buying power.
English
0
0
0
29
SPRT
SPRT@SPRT_RCCS·
RIKEN's Special Postdoctoral Researcher program goes into the next round, and my #HPC team is ready to hire you. You'll get topic flexibility, a small grant, and high salary with it, all while living in Kobe (beach and mountains included). riken.jp/en/careers/pro…
English
1
0
0
44
SPRT retweetledi
SCA/HPCAsia 2026
SCA/HPCAsia 2026@scahpcasia2026·
SCA/HPCAsia 2026 is just around the corner! Before coming to the venue, please read the Attendee Information👇 sca-hpcasia2026.jp/attend.html Registration desks are expected to be busy on the mornings of Jan 26 & 27. ⏰Early check-in is strongly recommended. #SCA_HPCAsia2026
SCA/HPCAsia 2026 tweet media
English
0
5
3
552
SPRT retweetledi
SPRT retweetledi
Satoshi Matsuoka
Satoshi Matsuoka@ProfMatsuoka·
And we win the Best Oaper Award with ORNL ! 3 out of 4 major conference awards! Amazing!
Satoshi Matsuoka tweet mediaSatoshi Matsuoka tweet mediaSatoshi Matsuoka tweet media
English
0
3
26
1.8K
SPRT
SPRT@SPRT_RCCS·
Join Ivan and Faveo's presentation on RAPTOR to explore the limits of low precision. The talk is schedulde for today at 3:30pm in room 274 at #SC25. if you miss it, check out the paper dl.acm.org/doi/10.1145/37…
SPRT tweet media
English
0
0
1
41
SPRT
SPRT@SPRT_RCCS·
Ivan Ivanov will present "Dynamic Thread Coarsening for CPU and GPU OpenMP Code" in the afternoon in the LLVM-HPC2025 workshop @Supercomputing . Join him in room 260.
English
0
0
0
33
SPRT retweetledi
RIKEN
RIKEN@riken_en·
*JAPAN SCIENTIST AI JAM SESSION* An event for Japan scientists to learn about the latest AI tools! Find out how AI can be applied to your own research! Staff from AWS, Anthropic, Google, and OpenAI will be there for support. trip.riken.jp/en/aijam/index…
RIKEN tweet media
English
0
1
1
1.4K
SPRT
SPRT@SPRT_RCCS·
Years ago we wrote a CUDA emulation layer for #Fugaku. And now I upgraded one of our @FujitsuHPC FX700 nodes with a real @nvidia GPU, and all seems to work. #FugakuNEXT
SPRT tweet media
English
0
1
2
104
SPRT
SPRT@SPRT_RCCS·
Now our GPU vendor is finally publicly announced. RIKEN R-CCS with co-design the next flagship supercomputer for Japan with the help of Nvidia and Fujitsu! Exciting 4 years ahead of us. blogs.nvidia.com/blog/fugakunex…
English
1
2
3
130
SPRT
SPRT@SPRT_RCCS·
Second rejection within a week because the editor or reviewer used AI to avoid thinking about our paper, and the LLM was likely some 1B phone model... 🤣 The system is broken beyond repair. over and out 🙊
English
1
0
0
44
SPRT
SPRT@SPRT_RCCS·
RAPTOR is accessible now, introducing a novel numerical profiler for your #HPC/science apps. You can get it on arxiv before the SC'25 release here arxiv.org/pdf/2507.04647. Faveo + Ivan did a fantastic job and are still improving the tool. Thanks to Anshu, Akash, Wahib, @wsmoses
SPRT tweet media
English
1
1
1
274
SPRT retweetledi
Satoshi Matsuoka
Satoshi Matsuoka@ProfMatsuoka·
Now that my ISCA 25 keynote commemorating Seymour Cray, I would like to solicit nominations for the IEEE-CS Cray Award. In fact the deadline is June 30th(!) computer.org/volunteering/a…
English
0
3
8
1.9K
SPRT
SPRT@SPRT_RCCS·
#HPC systems vs #algorithms advancements is kinda like "bandwidth vs latency". Scaling the former is bound (mostly) by money. Improving the latter has a fundamental limit of O(1). Luckily we have not reached the limits of either, so more progress will be coming 👍
English
0
0
1
28
SPRT
SPRT@SPRT_RCCS·
@elonmusk come to Japan. Great food, lots of mountains for your Boring company, thriving space and car/EV scene, and the best engineers. We'll help you build a few more Colossus, which are needed to reinvigorate the entrepreneurial spirit in the country. m.media-amazon.com/images/M/MV5BO…
English
0
0
0
17