HPC Guru (Twitter)
RT @HatemLtaief: It is indeed challenging to deploy such fine-grained mixed precision matrix algorithms on advanced architectures.
But why doing more flops when only a few are enough? Not to mention the saving in data movement and energy consumption.
Time to cease #oversolving.
#HPC https://t.co/014W4xDCxM
RT @HatemLtaief: It is indeed challenging to deploy such fine-grained mixed precision matrix algorithms on advanced architectures.
But why doing more flops when only a few are enough? Not to mention the saving in data movement and energy consumption.
Time to cease #oversolving.
#HPC https://t.co/014W4xDCxM
Twitter
Touisteur EmporteUneVache on X
@moyix @rohitgUCF Hearing @HatemLtaief talk about and explain dynamic precision adaptation in block/tile matrix operations, and some of the new (or recycled) ideas of whole-matrix mixed-precision accuracy adaptation in the industry made me rethink the whole…
HPCwire (Twitter)
Say It Loud: Kevin Hayden Is HPC Proud! As a Senior Network Engineer at Argonne National Laboratory, Kevin's dedication to the SC Conference spans two decades. Learn more about his valuable contributions and insights into the future of HPC. https://t.co/dD2WAV9M9t #iamhpc #HPC
Say It Loud: Kevin Hayden Is HPC Proud! As a Senior Network Engineer at Argonne National Laboratory, Kevin's dedication to the SC Conference spans two decades. Learn more about his valuable contributions and insights into the future of HPC. https://t.co/dD2WAV9M9t #iamhpc #HPC
HPCwire (Twitter)
Nominations are still open for the #HPCwire Readers' Choice Award until August 25, 2023 at 11:59 am PT. Don't miss your opportunity to make your voice heard! Help tell the story of #HPC in 2023. See the categories and make your nominations here: https://ow.ly/8kR150PtGGV
Nominations are still open for the #HPCwire Readers' Choice Award until August 25, 2023 at 11:59 am PT. Don't miss your opportunity to make your voice heard! Help tell the story of #HPC in 2023. See the categories and make your nominations here: https://ow.ly/8kR150PtGGV
HPC Guru (Twitter)
RT @EvanKirstel: Harvard researchers clone supercomputer on @googlecloud - @SiliconANGLE https://t.co/sZcR6dtnH1 thoughts @HPC_Guru?
RT @EvanKirstel: Harvard researchers clone supercomputer on @googlecloud - @SiliconANGLE https://t.co/sZcR6dtnH1 thoughts @HPC_Guru?
SiliconANGLE
Harvard researchers clone supercomputer on Google Cloud
Researchers at Harvard University have used Google LLC’s public cloud infrastructure platform to create a clone of a supercomputer that was used to perform a heart disease study.They claim it’
HPCwire (Twitter)
Primarily by using #CUDA (and #OpenCL, #OpenACC, etc.), #applications that are amenable to accelerated #processing have been optimized for #GPUs with great success. https://ow.ly/mjAk50PtGHR
Primarily by using #CUDA (and #OpenCL, #OpenACC, etc.), #applications that are amenable to accelerated #processing have been optimized for #GPUs with great success. https://ow.ly/mjAk50PtGHR
HPCwire (Twitter)
While the #qubit zoo keeps growing — #superconducting, trapped ions, neutral atoms, #photonics, spin qubits, and more — its members still share (at least so far) untenable error rates. https://ow.ly/OHtR50PtGHx
While the #qubit zoo keeps growing — #superconducting, trapped ions, neutral atoms, #photonics, spin qubits, and more — its members still share (at least so far) untenable error rates. https://ow.ly/OHtR50PtGHx
HPCwire (Twitter)
Important stakeholders and industry peers are eager to connect with you at HPC + AI Wall Street 2023. Have you registered yet? https://hpcaiwallstreet.com
#FinServ #FinTech #HPC #DataManagement #artificialintelligence #QuantumComputing #techconference #HPCAI2023
Important stakeholders and industry peers are eager to connect with you at HPC + AI Wall Street 2023. Have you registered yet? https://hpcaiwallstreet.com
#FinServ #FinTech #HPC #DataManagement #artificialintelligence #QuantumComputing #techconference #HPCAI2023
HPCwire (Twitter)
Introducing Jim Kovach, Director of Business Development in Higher Education and Research at PIER Group LLC. Jim has connected renowned institutions with cutting-edge computing solutions. Read more: https://ow.ly/K97Z50PtGIf #iamhpc #HPC #supercomputing
Introducing Jim Kovach, Director of Business Development in Higher Education and Research at PIER Group LLC. Jim has connected renowned institutions with cutting-edge computing solutions. Read more: https://ow.ly/K97Z50PtGIf #iamhpc #HPC #supercomputing
HPC Guru (Twitter)
RT @LorenaABarba: @HPC_Guru @EvanKirstel @googlecloud @SiliconANGLE @awscloud @Azure #Barbagroup students have been experimenting with #HPC workflows on the #cloud since it became barely feasible ~8 years ago.
In 2019: Reproducible workflow on Microsoft #Azure using containers and command-line configuration of the public cloud services…
https://t.co/bHn8pCz5a9
RT @LorenaABarba: @HPC_Guru @EvanKirstel @googlecloud @SiliconANGLE @awscloud @Azure #Barbagroup students have been experimenting with #HPC workflows on the #cloud since it became barely feasible ~8 years ago.
In 2019: Reproducible workflow on Microsoft #Azure using containers and command-line configuration of the public cloud services…
https://t.co/bHn8pCz5a9
HPC Guru (Twitter)
Dr. Amanda Randles is the 2023 ACM @SIGHPC Emerging Woman Leader in Technical Computing award winner
She holds over 120 patents, and her well-known blood flow simulation code, HARVEY, has been ported to the world’s top-ranked #supercomputers
Congratulations!
#HPC #SC23
Dr. Amanda Randles is the 2023 ACM @SIGHPC Emerging Woman Leader in Technical Computing award winner
She holds over 120 patents, and her well-known blood flow simulation code, HARVEY, has been ported to the world’s top-ranked #supercomputers
Congratulations!
#HPC #SC23
HPC Guru (Twitter)
SoftBank-owned chip designer @Arm has started the countdown to the biggest US IPO in almost two years
ft.com/content/36bc706f-66a1…
#HPC #AI #ARM
SoftBank-owned chip designer @Arm has started the countdown to the biggest US IPO in almost two years
ft.com/content/36bc706f-66a1…
#HPC #AI #ARM
Ft
SoftBank’s Arm unveils plans for biggest US IPO in nearly 2 years
Flotation of UK-based chip designer may help to revive market for primary listings after 18-month drought
HPC Guru (Twitter)
.@IBM watsonx #LLMs for code development can help migrate #COBOL applications to #Java
A key challenge with COBOL code is that there is a shrinking base of developers that actually know how to maintain it
venturebeat.com/ai/ibm-taps-…
#GenerativeAI #AI via @glennklockwood
.@IBM watsonx #LLMs for code development can help migrate #COBOL applications to #Java
A key challenge with COBOL code is that there is a shrinking base of developers that actually know how to maintain it
venturebeat.com/ai/ibm-taps-…
#GenerativeAI #AI via @glennklockwood
X (formerly Twitter)
#LLMs - Search / X
See posts about #LLMs on X. See what people are saying and join the conversation.
HPC Guru (Twitter)
Have to disagree with the illustrious @jimkxa here
#MooresLaw is not *only* about continuing improvements in the future, but there was a specific cadence at which we double transistors at iso-cost
Without the cadence, IMHO, the law (as I understand it) is dead
#HPC #AI
Have to disagree with the illustrious @jimkxa here
#MooresLaw is not *only* about continuing improvements in the future, but there was a specific cadence at which we double transistors at iso-cost
Without the cadence, IMHO, the law (as I understand it) is dead
#HPC #AI
HPC Guru (Twitter)
RT @thoefler: Join us tomorrow (Wed) online at 1:50pm PST for the Panel "EtherNET vs EtherNOT … again?"
Teaser: @ultraethernet is the new kid on the block and on the rise.
hoti.org/program/ - join via zoom, register for free now. More than 500 attendees already!
#HPC #AI #datacenter
RT @thoefler: Join us tomorrow (Wed) online at 1:50pm PST for the Panel "EtherNET vs EtherNOT … again?"
Teaser: @ultraethernet is the new kid on the block and on the rise.
hoti.org/program/ - join via zoom, register for free now. More than 500 attendees already!
#HPC #AI #datacenter
HotI30 (2023)
Program
ProgramAll times are in UTC -7 (America/Los_Angeles) Day 1 (Aug, 23rd) Day 2 (Aug, 24th) Day 3 (Aug, 25th) Day 1 (Aug, 23rd) Time (UTC-7) Title 09:00 Introduction and WelcomeSession Chair: Taylor Groves (Lightmatter) 09:15
HPC Guru (Twitter)
#MPI Application Binary Interface Standardization
MPI is the most widely used interface for #HPC workloads
A new level of MPI compatibility: a standard #ABI is proposed in this paper
arxiv.org/pdf/2308.11214.pdf
via @science_dot
#MPI Application Binary Interface Standardization
MPI is the most widely used interface for #HPC workloads
A new level of MPI compatibility: a standard #ABI is proposed in this paper
arxiv.org/pdf/2308.11214.pdf
via @science_dot
HPC Guru (Twitter)
RT @dnystedt: Nvidia will at least triple the production of its top H100 AI processor in 2024 to 1.5mn to 2mn H100s, a massive jump from the 500,000 expected this year, the Financial Times reports, noting the AI processors are already sold out into 2024. $NVDA $TSM
ft.com/content/c7e9cfa9-3f68…
RT @dnystedt: Nvidia will at least triple the production of its top H100 AI processor in 2024 to 1.5mn to 2mn H100s, a massive jump from the 500,000 expected this year, the Financial Times reports, noting the AI processors are already sold out into 2024. $NVDA $TSM
ft.com/content/c7e9cfa9-3f68…
Twitter
Dan Nystedt on X
Nvidia will at least triple the production of its top H100 AI processor in 2024 to 1.5mn to 2mn H100s, a massive jump from the 500,000 expected this year, the Financial Times reports, noting the AI processors are already sold out into 2024. $NVDA $TSM
h…
h…
HPC Guru (Twitter)
2nd gen GH200 superchip: @nvidia is moving Hopper to #HBM3e to boost capacity to 141 GB & bandwidth to 5 TB/s
Compared to the original Hopper SXM5 device announced last year:
Capacity: +76.3%
Bandwidth: +49.3 %
nextplatform.com/2023/08/22/…
#HPC #AI #LLM via @TDaytonPM
2nd gen GH200 superchip: @nvidia is moving Hopper to #HBM3e to boost capacity to 141 GB & bandwidth to 5 TB/s
Compared to the original Hopper SXM5 device announced last year:
Capacity: +76.3%
Bandwidth: +49.3 %
nextplatform.com/2023/08/22/…
#HPC #AI #LLM via @TDaytonPM
HPC Guru (Twitter)
NCAR inaugurated the new Derecho #supercomputer on Aug 18 with a ceremonial ribbon cutting
Derecho is a @HPE_Cray system based on @AMDServer Milan CPUs and some @nvidia A100 GPUs
news.ucar.edu/132907/officia…
#HPC #AI via @NCAR_Science
NCAR inaugurated the new Derecho #supercomputer on Aug 18 with a ceremonial ribbon cutting
Derecho is a @HPE_Cray system based on @AMDServer Milan CPUs and some @nvidia A100 GPUs
news.ucar.edu/132907/officia…
#HPC #AI via @NCAR_Science
HPC Guru (Twitter)
#AI chip startup @EsperantoTech pivoted its focus from recommendation acceleration to #LLM inference & #HPC
Released a SDK and PCIe accelerator card for its ET-SoC-1 first generation RISC-V accelerator chip
eetimes.com/esperanto-pivots…
#RISCV via @sallywf
#AI chip startup @EsperantoTech pivoted its focus from recommendation acceleration to #LLM inference & #HPC
Released a SDK and PCIe accelerator card for its ET-SoC-1 first generation RISC-V accelerator chip
eetimes.com/esperanto-pivots…
#RISCV via @sallywf
X (formerly Twitter)
#AI - Search / X
See posts about #AI on X. See what people are saying and join the conversation.
HPC Guru (Twitter)
RT @s_thiell: Today we reached a total of 4.9 PB of flash storage at 200 GB/s for @StanfordCompute’s Sherlock to support a wide range of research workloads. Lustre 2.15. #HPC #AI
RT @s_thiell: Today we reached a total of 4.9 PB of flash storage at 200 GB/s for @StanfordCompute’s Sherlock to support a wide range of research workloads. Lustre 2.15. #HPC #AI