English
English Indonesia Türkçe Deutsch Español Português Pусский Français Italiano Nederlands Polski العربية ไทย Tiếng Việt 繁體中文 简体中文 日本語 한국어
Geekpie_HPC's profile picture. Geek hub for seeking infinity. From @ShanghaiTechUni

GeekPie_HPC

@Geekpie_HPC

Geek hub for seeking infinity. From @ShanghaiTechUni

117Posts 59Followers 112Following
You might like
GeekPie_HPC reposted
hpccouncil's profile picture.
HPC-AI Advisory Council
 @hpccouncil
May 26, 2023

👏Congratulations to our remote teams who participated in the 2023 Student Cluster Competition. - First Place: NTU, @realntuhpc, Singapore - Second Place: UCSC, @notsoslowslugs, USA - Third Place: ShanghaiTech, @geekpie_hpc, China #HPC #imaginetomorrow #ISC23_SCC

0
3
7
0
554
Geekpie_HPC's profile picture.
GeekPie_HPC
 @Geekpie_HPC
May 3, 2023

Stickers will be available @aschpc @InspurCorp And thanks a lot @InspurCorp for sponsoring. 🥰

Geekpie_HPC's tweet image. Stickers will be available @aschpc @InspurCorp And thanks a lot @InspurCorp for sponsoring. 🥰
0
0
4
0
203
Geekpie_HPC's profile picture.
GeekPie_HPC
 @Geekpie_HPC
Apr 28, 2023

Vote for us plz 🥰

aschpc's profile picture. Asia Supercomputer Community The key events of the ASC: 1 ASC Student Supercomputer Challenge 2 AICC&HPCUF 3 HPC Connection Workshop
ASC HPC Challenge
 @aschpc
Apr 28, 2023

(4/6) Pick your favorite team for the "Most Popular Team" title! Two rounds of votes will take place before May 10th. The clock is ticking.⏰Vote now! #ASC22

0
0
0
0
845
0
2
2
0
284
GeekPie_HPC reposted
DeemosTech's profile picture.
Deemos
@DeemosTech
Apr 25, 2023

Successfully generated 3D assets with PBR materials from AI-generated or handcrafted concept art with multiple angles. This image-to-3D pipeline will be a part of hyperhuman.top in the future! Text->SD->Hyperhuman will be your new workflow! #GenerativeAI…

DeemosTech's tweet image. Successfully generated 3D assets with PBR materials from AI-generated or handcrafted concept art with multiple angles. This image-to-3D pipeline will be a part of hyperhuman.top in the future! Text->SD->Hyperhuman will be your new workflow! #GenerativeAI…
DeemosTech's tweet image. Successfully generated 3D assets with PBR materials from AI-generated or handcrafted concept art with multiple angles. This image-to-3D pipeline will be a part of hyperhuman.top in the future! Text->SD->Hyperhuman will be your new workflow! #GenerativeAI…
1
59
263
109
37K
Geekpie_HPC's profile picture.
GeekPie_HPC
 @Geekpie_HPC
Apr 21, 2023

See you @ USTC

aschpc's profile picture. Asia Supercomputer Community The key events of the ASC: 1 ASC Student Supercomputer Challenge 2 AICC&HPCUF 3 HPC Connection Workshop
ASC HPC Challenge
 @aschpc
Apr 21, 2023

Check out the agenda for the 10th #ASC Student Supercomputer Challenge! The contestants will compete on the most exciting topics in super #computing and beyond, including #LLM, DeePMD, etc. Looking forward to meeting these talented teams! Learn more: asc-events.net

aschpc's tweet image. Check out the agenda for the 10th #ASC Student Supercomputer Challenge! The contestants will compete on the most exciting topics in super #computing and beyond, including #LLM, DeePMD, etc. Looking forward to meeting these talented teams! Learn more: asc-events.net
0
0
5
0
399
0
0
1
0
155
GeekPie_HPC reposted
spackpm's profile picture.
Spack
 @spackpm
Apr 20, 2023

Spack v0.19.2 is out! This is a bugfix release. Some highlights: - Fixes for using variants in requirements - Fix issues with compiler detection on macOS - Improved stability for parallel installs - more! Details at github.com/spack/spack/re…

github.com

Release v0.19.2 · spack/spack

v0.19.2 (2023-04-04) Spack Bugfixes Ignore global variant requirement for packages that do not define it (#35037) Compiler wrapper: improved parsing of linker arguments (#35929, #35912) Do not det...

Source: github.com
0
4
10
0
979
GeekPie_HPC reposted
thoefler's profile picture.
Torsten Hoefler 🇨🇭
 @thoefler
Apr 15, 2023

I am happy to confirm that Aurora looks and sounds complete (still noisy despite the water cooling). Last time I saw it at @argonne a year ago, photos were not ok yet ;-). The 2nd exaflop computer at an expected cost of $500 million running #HPC and #AI! en.wikipedia.org/wiki/Aurora_(s…

thoefler's tweet image. I am happy to confirm that Aurora looks and sounds complete (still noisy despite the water cooling). Last time I saw it at @argonne a year ago, photos were not ok yet ;-). The 2nd exaflop computer at an expected cost of $500 million running #HPC and #AI! en.wikipedia.org/wiki/Aurora_(s…
thoefler's tweet image. I am happy to confirm that Aurora looks and sounds complete (still noisy despite the water cooling). Last time I saw it at @argonne a year ago, photos were not ok yet ;-). The 2nd exaflop computer at an expected cost of $500 million running #HPC and #AI! en.wikipedia.org/wiki/Aurora_(s…
5
32
177
9
25K
GeekPie_HPC reposted
murezsy's profile picture.
🦦🚰🏴‍☠️🌈
 @murezsy
Apr 13, 2023

？（QAQ

Supercomputing's profile picture. Official Twitter for the SC Conference Series • SC25 • Nov 16–21, 2025 • America’s Center, St. Louis, MO
SC25
 @Supercomputing
Apr 13, 2023

Get inspired by the best in the field! Registration for #SIGGRAPH2023 is now open 🎉. Join @siggraph, the premier conference for computer graphics and interactive techniques, in LA from 6-10 August. Register by 28 April for early bird pricing! buff.ly/3ml9XfQ

Supercomputing's tweet image. Get inspired by the best in the field! Registration for #SIGGRAPH2023 is now open 🎉. Join @siggraph, the premier conference for computer graphics and interactive techniques, in LA from 6-10 August. Register by 28 April for early bird pricing! buff.ly/3ml9XfQ
0
0
3
0
1K
1
1
1
0
154
GeekPie_HPC reposted
ernstdj's profile picture.
Dan Ernst
 @ernstdj
Apr 6, 2023

The key technologies enabling CXL to grow will be entirely about performance management. Technologies is plural because the performance context is different for everybody.

This post is unavailable.
0
1
1
0
452
GeekPie_HPC reposted
JeffDean's profile picture.
Jeff Dean
@JeffDean
Apr 5, 2023

Paper: TPUv4 system has an optically reconfigurable network to assemble groups of 4x4x4 chips like legos (4x4x12? 16x16x16?). SparseCores help w/ embeddings. TPUv4 outperforms TPUv3 by 2.1x & perf/W by 2.7x, & has 4096 chips so ~10x faster overall. .arxiv.org/abs/2304.01433#

11
115
609
86
158K
GeekPie_HPC reposted
thoefler's profile picture.
Torsten Hoefler 🇨🇭
 @thoefler
Apr 3, 2023

Are you also rushing towards the @Supercomputing deadline this week? Keep good scientific practice for #HPC benchmarking in mind. Following the 12 "scientific benchmarking" rules (htor.inf.ethz.ch/publications/i…) will help to improve your paper :-). Good luck to all submitters to #SC23

thoefler's tweet image. Are you also rushing towards the @Supercomputing deadline this week? Keep good scientific practice for #HPC benchmarking in mind. Following the 12 "scientific benchmarking" rules (htor.inf.ethz.ch/publications/i…) will help to improve your paper :-). Good luck to all submitters to #SC23
thoefler's tweet image. Are you also rushing towards the @Supercomputing deadline this week? Keep good scientific practice for #HPC benchmarking in mind. Following the 12 "scientific benchmarking" rules (htor.inf.ethz.ch/publications/i…) will help to improve your paper :-). Good luck to all submitters to #SC23
thoefler's tweet image. Are you also rushing towards the @Supercomputing deadline this week? Keep good scientific practice for #HPC benchmarking in mind. Following the 12 "scientific benchmarking" rules (htor.inf.ethz.ch/publications/i…) will help to improve your paper :-). Good luck to all submitters to #SC23
thoefler's tweet image. Are you also rushing towards the @Supercomputing deadline this week? Keep good scientific practice for #HPC benchmarking in mind. Following the 12 "scientific benchmarking" rules (htor.inf.ethz.ch/publications/i…) will help to improve your paper :-). Good luck to all submitters to #SC23
1
13
49
13
8K
GeekPie_HPC reposted
msbaz2013's profile picture.
Muhammad Shahbaz
 @msbaz2013
Mar 28, 2023

Tushar Swamy presenting Homunculus at @ASPLOSConf '23 today! @PurdueCS | @Stanford

msbaz2013's tweet image. Tushar Swamy presenting Homunculus at @ASPLOSConf '23 today! @PurdueCS | @Stanford
msbaz2013's tweet image. Tushar Swamy presenting Homunculus at @ASPLOSConf '23 today! @PurdueCS | @Stanford
msbaz2013's tweet image. Tushar Swamy presenting Homunculus at @ASPLOSConf '23 today! @PurdueCS | @Stanford
msbaz2013's tweet image. Tushar Swamy presenting Homunculus at @ASPLOSConf '23 today! @PurdueCS | @Stanford
msbaz2013's profile picture. Assistant Professor of Computer Science and Engineering (CSE) at the University of Michigan
Muhammad Shahbaz
 @msbaz2013
Feb 8, 2023

Another step toward making *In-network ML* a reality. Our upcoming paper, Homunculus, at @ASPLOSConf '23 presents a framework for network operators to automatically generate ML data-plane pipelines without requiring expertise in ML model development and switch architectures!

msbaz2013's tweet image. Another step toward making *In-network ML* a reality. Our upcoming paper, Homunculus, at @ASPLOSConf '23 presents a framework for network operators to automatically generate ML data-plane pipelines without requiring expertise in ML model development and switch architectures!
1
5
39
4
6K
0
5
20
2
4K
GeekPie_HPC reposted
IanCutress's profile picture.
𝐷𝑟. 𝐼𝑎𝑛 𝐶𝑢𝑡𝑟𝑒𝑠𝑠
@IanCutress
Mar 13, 2023

Congrats to the new @AMD Fellows! - Ralph Wittig (Xilinx AI) - Ben Sander (Opteron, ROCm) - Suresh Ramalingam (2.5D/3D) - Nathan Kalyanasundharam (IF/CXL/UCIe) - David Blythe (DX, OpenGL) AMD now has 13 corporate fellows.

1
14
141
3
21K
GeekPie_HPC reposted
Underfox3's profile picture.
Underfox
 @Underfox3
Feb 25, 2023

In this paper is presented MPIWasm, a novel Wasm embedder for MPI-based #HPC applications that enables high-performance execution of Wasm code, has low-overhead for MPI calls, and supports high-performance networking interconnects present on HPC systems. arxiv.org/pdf/2301.03982…

Underfox3's tweet image. In this paper is presented MPIWasm, a novel Wasm embedder for MPI-based #HPC applications that enables high-performance execution of Wasm code, has low-overhead for MPI calls, and supports high-performance networking interconnects present on HPC systems. arxiv.org/pdf/2301.03982…
Underfox3's tweet image. In this paper is presented MPIWasm, a novel Wasm embedder for MPI-based #HPC applications that enables high-performance execution of Wasm code, has low-overhead for MPI calls, and supports high-performance networking interconnects present on HPC systems. arxiv.org/pdf/2301.03982…
1
5
18
1
2K
GeekPie_HPC reposted
DrHaiAhNam's profile picture.
Dr. Hai Ah Nam
 @DrHaiAhNam
Feb 16, 2023

There's nothing like being in a room of people who get you. If you think developer productivity and software sustainability should be a priority to scientific research, then you're in the right place. ❤️

BetterSSw's profile picture. BSSw supports people, practices, techniques, experiences, and tools to improve CSE software productivity, quality and sustainability.
Better Scientific Software (BSSw)
 @BetterSSw
Feb 16, 2023

Today (2/16) is the last day of the @exascaleproject Community BoF Days. Join the #BSSw Fellowship BOF @ 1PM ET if you're passionate about quality scientific software. Free but must register. exascaleproject.org/event/2023-ecp… #somycodewillseethefuture

exascaleproject.org

2023 ECP Community BOF Days

When: February 14, 2023 – February 16, 2023 all-day The Exascale Computing Project (ECP) 2023 Community Birds-of-a-Feather (BOF) Days will take place February 14–16, with multiple sessions each day....

Source: exascaleproject.org
0
0
1
1
2K
0
6
16
0
2K
GeekPie_HPC reposted
HPC_Guru's profile picture.
HPC Guru
 @HPC_Guru
Feb 16, 2023

Intel Publishes Blazing Fast AVX-512 Sorting Library The code has been merged to Numpy and is providing some 10~17x speed-ups On an Intel Tigerlake system this sped-up 16-bit int sorting by 17x phoronix.com/news/Intel-AVX… #HPC #AI via @HPCZorf @phoronix

HPC_Guru's tweet card. Intel recently published an open-source C++ header file library for high performance SIMD-based sorting, which initially is focused on providing a lightning fast AVX-512 quicksort implementation
Intel Publishes Blazing Fast AVX-512 Sorting Library, Numpy Switching To It For 10~17x Faster Sorts
Source: phoronix.com
5
28
120
13
41K

United States Trends

You might like

Something went wrong.


Something went wrong.