HPC Guru (Twitter)
RT @ProfMatsuoka: @ProfMatsuoka @HPC_Guru @nvidia @dylan522p: In fact training GPU market will saturate soon, as the power law in training connotes brute force compute advances to next gen models is challenging given all hyperscalars are maxed out using Exascale GPUs supercomputers already. But inference will grow proportionally to usage of LLMs
RT @ProfMatsuoka: @ProfMatsuoka @HPC_Guru @nvidia @dylan522p: In fact training GPU market will saturate soon, as the power law in training connotes brute force compute advances to next gen models is challenging given all hyperscalars are maxed out using Exascale GPUs supercomputers already. But inference will grow proportionally to usage of LLMs
HPC Guru (Twitter)
RT @NatLabsHPC: Curious about what's going on at this year's DOE National Labs booth at @Supercomputing #SC23?
Check out our featured speakers, demos, and more at: scdoe.info/
RT @NatLabsHPC: Curious about what's going on at this year's DOE National Labs booth at @Supercomputing #SC23?
Check out our featured speakers, demos, and more at: scdoe.info/
HPC Guru (Twitter)
Countdown to #SC23: @insideHPC assembled a panel to recommend what to attend and trends to watch during @Supercomputing in Denver
o @loiscmc
o @sunitachandra29
o @addisonsnell
o Earl Joseph @HPC_Hyperion
youtube.com/watch?v=jYjo7Moa…
#HPC #AI #IamHPC
Countdown to #SC23: @insideHPC assembled a panel to recommend what to attend and trends to watch during @Supercomputing in Denver
o @loiscmc
o @sunitachandra29
o @addisonsnell
o Earl Joseph @HPC_Hyperion
youtube.com/watch?v=jYjo7Moa…
#HPC #AI #IamHPC
HPC Guru (Twitter)
.@nvidia is prepping three new #GPUs for #AI and #HPC applications tailored for China and to comply with U.S. export requirements
o HGX H20
o L20 PCle
o L2 PCle
tomshardware.com/tech-indust…
#GPU via @tomshardware
.@nvidia is prepping three new #GPUs for #AI and #HPC applications tailored for China and to comply with U.S. export requirements
o HGX H20
o L20 PCle
o L2 PCle
tomshardware.com/tech-indust…
#GPU via @tomshardware
HPC Guru (Twitter)
Human:1 GPT:0
@gcs_hpc_guru's limerick is better than what was generated by #GPT4
The HPC Guru of Twitter
Whose tweets are by no means just litter
Once flew into Denver
Around mid-November
And no one could flush out that critter
#HPC #AI #SC23 #IMangoHPC
Human:1 GPT:0
@gcs_hpc_guru's limerick is better than what was generated by #GPT4
The HPC Guru of Twitter
Whose tweets are by no means just litter
Once flew into Denver
Around mid-November
And no one could flush out that critter
#HPC #AI #SC23 #IMangoHPC
HPC Guru (Twitter)
GENCI/CEA, FZJ, and PASQAL announce progress in the framework of the European project High-Performance Computer and Quantum Simulator hybrid (#HPCQS)
hpcwire.com/off-the-wire/gen…
#HPC #QuantumComputing #SC23 via @HPCwire
GENCI/CEA, FZJ, and PASQAL announce progress in the framework of the European project High-Performance Computer and Quantum Simulator hybrid (#HPCQS)
hpcwire.com/off-the-wire/gen…
#HPC #QuantumComputing #SC23 via @HPCwire
HPC Guru (Twitter)
RT @LBNLcs: DYK @UCBerkeley & @BerkeleyLab's @KathyYelick will participate in 2 events @Supercomputing on Nov. 13?
She'll give the @womeninHPC keynote at 9 a.m. MST and will join #SC23's opening plenary, “I Am HPC: Impact and Future Directions” at 5:30 p.m. MST bit.ly/SC23ky #HPC
RT @LBNLcs: DYK @UCBerkeley & @BerkeleyLab's @KathyYelick will participate in 2 events @Supercomputing on Nov. 13?
She'll give the @womeninHPC keynote at 9 a.m. MST and will join #SC23's opening plenary, “I Am HPC: Impact and Future Directions” at 5:30 p.m. MST bit.ly/SC23ky #HPC
HPC Guru (Twitter)
RT @IanCutress: New @HPC_Guru avatar.
Who is going to be the first to commission plushies? Do we have enough time before SC23?
RT @IanCutress: New @HPC_Guru avatar.
Who is going to be the first to commission plushies? Do we have enough time before SC23?
HPC Guru (Twitter)
.@MicronTech announces 32Gb monolithic die-based 128GB #DDR5 RDIMM #memory with performance of up to 8000 MT/s to support data center workloads
investors.micron.com/news-re…
#HPC #AI #GenAI
.@MicronTech announces 32Gb monolithic die-based 128GB #DDR5 RDIMM #memory with performance of up to 8000 MT/s to support data center workloads
investors.micron.com/news-re…
#HPC #AI #GenAI
HPC Guru (Twitter)
SPEC announces the SPECaccel 2023 benchmark suite
It measures the performance of key system components using computationally intensive parallel applications running under the #OpenACC and #OpenMP target offloading APIs
spec.org/accel2023/
#HPC #AI
SPEC announces the SPECaccel 2023 benchmark suite
It measures the performance of key system components using computationally intensive parallel applications running under the #OpenACC and #OpenMP target offloading APIs
spec.org/accel2023/
#HPC #AI
HPC Guru (Twitter)
RT @NVIDIADC: Join speakers from @nvidia, @fz_juelich, @cscsch, and @TACC as they discuss the latest in scientific computing on November 13 during NVIDIA's #SC23 Special Address. @Supercomputing
nvda.ws/3QRXKeU
RT @NVIDIADC: Join speakers from @nvidia, @fz_juelich, @cscsch, and @TACC as they discuss the latest in scientific computing on November 13 during NVIDIA's #SC23 Special Address. @Supercomputing
nvda.ws/3QRXKeU
HPC Guru (Twitter)
RT @Livermore_Comp: What time is it? ⏰ Time to start posting about #SC23. See you next week!
computing.llnl.gov/sc23-even…
#LLNLatSC #HPC #computing
@Livermore_Lab @LLNL_OpenSource @NatLabsHPC
RT @Livermore_Comp: What time is it? ⏰ Time to start posting about #SC23. See you next week!
computing.llnl.gov/sc23-even…
#LLNLatSC #HPC #computing
@Livermore_Lab @LLNL_OpenSource @NatLabsHPC
HPCwire (Twitter)
Core42 Sets New Benchmark for Arabic Large Language Models with the Release of Jais 30B
ow.ly/7TnI50Q615J @G42ai @core42_ai #HPC #TCIwire
Core42 Sets New Benchmark for Arabic Large Language Models with the Release of Jais 30B
ow.ly/7TnI50Q615J @G42ai @core42_ai #HPC #TCIwire
HPCwire
Core42 Sets New Benchmark for Arabic Large Language Models with the Release of Jais 30B
ABU DHABI, United Arab Emirates, Nov. 9, 2023 — Core42, a G42 company and the UAE-based national-scale enabler for cloud and generative AI, announced the launch of Jais 30B, the […]
HPCwire (Twitter)
SPEC Delivers Major Update to SPECaccel Benchmark Suite
ow.ly/ovfm50Q6828 @spec_perf #HPC #TCIwire
SPEC Delivers Major Update to SPECaccel Benchmark Suite
ow.ly/ovfm50Q6828 @spec_perf #HPC #TCIwire
HPCwire
SPEC Delivers Major Update to SPECaccel Benchmark Suite
GAINESVILLE, Va., Nov. 9, 2023 — The Standard Performance Evaluation Corporation (SPEC), a trusted global leader in computing benchmarks, today announced the availability of the SPECaccel 2023 benchmark suite, a major […]
HPCwire (Twitter)
Intel Gaudi AI Accelerator Gains 2x Performance Leap on GPT-3 with FP8 Software
ow.ly/fsAK50Q64K2 @intel #HPC #TCIwire
Intel Gaudi AI Accelerator Gains 2x Performance Leap on GPT-3 with FP8 Software
ow.ly/fsAK50Q64K2 @intel #HPC #TCIwire
HPCwire
Intel Gaudi AI Accelerator Gains 2x Performance Leap on GPT-3 with FP8 Software
Nov. 9, 2023 — MLCommons published results of the industry standard MLPerf training v3.1 benchmark for training AI models, with Intel submitting results for Intel Gaudi2 accelerators and 4th Gen […]
HPC Guru (Twitter)
RT @glennklockwood: Alternative headline: Intel begins divesting itself of DAOS by transferring liabilities to a consortium analogous to OpenSFS for Lustre.
Good for the community, but I wonder how the Aurora folks feel about this.
RT @glennklockwood: Alternative headline: Intel begins divesting itself of DAOS by transferring liabilities to a consortium analogous to OpenSFS for Lustre.
Good for the community, but I wonder how the Aurora folks feel about this.
HPC Guru (Twitter)
RT @dnystedt: SK Hynix plans to invest ₩10 trillion won (US$7.6 billion) on memory chip facilities in 2024 as it ramps production of advanced memory chips, HBM3, DDR5 and LPDDR5, and TSV packaging, amid competition with Samsung, media report, saying the 2024 capex figure is up versus ₩6 trillion to ₩7 trillion this year. #skhynix #Samsung $MU #semiconductors kedglobal.com/korean-chipmak…
RT @dnystedt: SK Hynix plans to invest ₩10 trillion won (US$7.6 billion) on memory chip facilities in 2024 as it ramps production of advanced memory chips, HBM3, DDR5 and LPDDR5, and TSV packaging, amid competition with Samsung, media report, saying the 2024 capex figure is up versus ₩6 trillion to ₩7 trillion this year. #skhynix #Samsung $MU #semiconductors kedglobal.com/korean-chipmak…
HPC Guru (Twitter)
RT @nicholasmalaya: Looking forward to #SC23 this year. I'll be attending the technical program w/ two talks:
"The MI300 APU: Programming for CPUs and GPUs on a single package" nowlab.cse.ohio-state.edu/es…
"Experiences Readying Applications for Exascale" (supercomputing.org)
See you there!
RT @nicholasmalaya: Looking forward to #SC23 this year. I'll be attending the technical program w/ two talks:
"The MI300 APU: Programming for CPUs and GPUs on a single package" nowlab.cse.ohio-state.edu/es…
"Experiences Readying Applications for Exascale" (supercomputing.org)
See you there!
HPC Guru (Twitter)
RT @ProfMatsuoka: At U-Tokyo Supercomputing Center (ITC) today, touring BDEC-1 machine, half of which is an offspring of #Fugaku, the other half being a GPU machine.
RT @ProfMatsuoka: At U-Tokyo Supercomputing Center (ITC) today, touring BDEC-1 machine, half of which is an offspring of #Fugaku, the other half being a GPU machine.