What do IKEA manuals & algorithms have in common? Both break complex tasks into simple steps! π οΈ
@eurohpc-ju.bsky.social @inno4scale.bsky.social drives algorithmic innovation for Europe's #exascale #supercomputers
πwww.eurohpc-ju.europa.eu/next-level-s...
20.11.2025 11:12 β
π 3
π 2
π¬ 0
π 1
Inno4scale Team in a meeting room
Luxembourg sign at the airport
Inno4scale Team in a meeting room
Inno4scale partners met today in Luxembourg for the projectβs final review meeting with the @eurohpc-ju.bsky.social project officer and reviewers.
We appreciated the constructive and positive feedback and look forward to the final report.
Thanks to all involved for the strong collaboration!
10.09.2025 18:06 β
π 1
π 0
π¬ 0
π 0
πHave you checked this out?
The @inno4scale.bsky.social Success Story Highlights Booklet is available online!
Explore 22 innovation studies showing how scalable #HPC enables breakthroughs in science, industry & public administration!
ππRead more: www.inno4scale.eu/success-stor...
10.09.2025 13:53 β
π 3
π 1
π¬ 0
π 0
Cover of the Inno4scale Success Story Highlights Booklet
π New from Inno4scale: A booklet with selected highlights from 22 innovation studies on algorithms for European exascale systems.
Scalable solvers, mixed-precision CFD, sparse matrix kernels & more.
Check it out β‘οΈhttps://www.inno4scale.eu/success-stories/
28.07.2025 06:50 β
π 2
π 1
π¬ 0
π 0
Matrix sparsity hides useful patterns for solver choice. Discuss the in the comments π‘π§
π¬ What does the LimitX recommender predict from sparse matrix structure?
1οΈβ£ Solver memory usage
2οΈβ£ Spectral matrix properties
3οΈβ£ Eigenvector orientations
Look for hints ππ www.inno4scale.eu
18.06.2025 09:34 β
π 0
π 0
π¬ 0
π 0
Weβre proud to share that Inno4scale was presented as part of the EuroHPC Joint Undertaking booth at this yearβs ISC High Performance Conference in Hamburg! π
A big thank you to everyone who stopped by the booth to learn more about Inno4scale, share ideas, and engage in insightful discussions π§ π₯οΈ
16.06.2025 08:51 β
π 1
π 0
π¬ 0
π 0
GPU stalls limit performance in exascale CFD solvers. Discuss the question below in the comments π‘π§
π¬ What limits GPU scalability in coarse-grid solvers?
1οΈβ£ Too few elements per core
2οΈβ£ CPU driven communication stalls
3οΈβ£ Excessive matrix assembly costs
Look for hints ππ www.inno4scale.eu
11.06.2025 09:34 β
π 0
π 0
π¬ 0
π 0
Two way EL simulations struggle to scale due to stepwise synchronisation. SCALETRACK changes that. Discuss the following question in the comments π‘π§
π¬ What enables scalable EL simulations?
1οΈβ£ Particle preaveraging
2οΈβ£ Async coupling
3οΈβ£ Interpolation caching
Look for hints π www.inno4scale.eu
04.06.2025 09:34 β
π 0
π 0
π¬ 0
π 0
Sparse training for LLMs can reduce costs and emissions but hardware limits still apply. A new approach changes that. Discuss in comments on the question below π‘π§
π¬ What enables full-range sparsity on GPUs?
1οΈβ£ Mixed precision
2οΈβ£ VENOM format
3οΈβ£ cuSparse
Look for hints π www.inno4scale.eu
28.05.2025 09:34 β
π 0
π 0
π¬ 0
π 0
GNNs struggle with memory and speed. CBM4Scale compresses binary matrices by reusing similar rows. Discuss in the comments on the following question π‘π§
π¬ Which approach supports performance portability?
1οΈβ£ OpenACC
2οΈβ£ Domain-specific libs
3οΈβ£ Manual CUDA
Look for hints ππ www.inno4scale.eu
21.05.2025 09:34 β
π 0
π 0
π¬ 0
π 0
π New on the Inno4scale Blog
Weβve wrapped up the Inno4scale project with a the Inno4scale Innovation Day in Barcelona!
Our latest article offers an overview on the final event π°
π Read the full article: lnkd.in/e9Uds7GD
#HPC #Exascale #AI #Supercomputing #ScientificComputing #MareNostrum
16.05.2025 07:21 β
π 1
π 0
π¬ 0
π 0
Donβt miss Inno4scale Innovation Day!
ποΈMay 13, 2025 | 10:30β17:30 CEST
πBarcelona
Watch livestream: shorturl.at/Rh3vr
#Inno4scale #EuroHPC #Quantum
12.05.2025 09:22 β
π 3
π 2
π¬ 0
π 0
Take yourβlast chanceβ to be part of the Inno4scale Innovation Day tomorrow via YouTube livestream! π
πΊ Join the livestream here: www.youtube.com/live/M1X3JOn...
#HPC #Exascale #Supercomputing #Innovation #Inno4scale #Research #Collaboration #Barcelona
12.05.2025 09:01 β
π 0
π 0
π¬ 0
π 0
Inno4scale Innovation Day
YouTube video by Inno4scale
Reminder π‘
The Inno4scale Innovation Day on May 13 goes hybrid and will also be available via YouTube livestream! π
πΊ Join the livestream here: www.youtube.com/live/M1X3JOn...
#HPC #Exascale #Supercomputing #Innovation #Inno4scale #Research #Collaboration #Barcelona
09.05.2025 07:29 β
π 1
π 0
π¬ 0
π 0
βΌοΈExciting updateβΌοΈ
Inno4scale Innovation Day goes hybrid! π‘
If you can't join us onsite in Barcelona, then watch the livestream on May 13 π
πΊ rtmp://a.rtmp.youtube.com/live2
02.05.2025 08:54 β
π 1
π 0
π¬ 0
π 0
βΌοΈLast chanceβΌοΈ
π‘Don't forget to join the Inno4scale Innovation Day on May 13 in Barcelona π
This event marks the successful conclusion of the Inno4scale project and provides a platform for future collaborations in HPC π₯οΈπ©βπ¬
Registrationπ regi.hlrs.de/2025/inno4sc...
24.04.2025 09:30 β
π 2
π 0
π¬ 0
π 1
ML speeds up atomic simulations, but key challenges remain. Discuss in the comments! π‘π§
π¬ What is the main challenge for MLPs?
1οΈβ£ Slow computation
2οΈβ£ Missing long-range effects
3οΈβ£ Limited training data
Look for hints ππ www.inno4scale.eu
23.04.2025 06:01 β
π 0
π 0
π¬ 0
π 0
βReminderβ
π‘Don't forget to join the Inno4scale Innovation Day on May 13 in Barcelona π
This event marks the successful conclusion of the Inno4scale project and provides a platform for future collaborations in HPC π₯οΈπ©βπ¬
Registrationπ regi.hlrs.de/2025/inno4sc...
22.04.2025 09:30 β
π 1
π 0
π¬ 0
π 0
HPC speeds up simulations, but sync barriers slow scaling. Discuss in the comments! π‘π§
π¬ What is the main advantage of AMCG?
1οΈβ£ Reduces accuracy
2οΈβ£ Increases memory
3οΈβ£ No sync barriers
Look for hints ππ www.inno4scale.eu
17.04.2025 06:01 β
π 0
π 0
π¬ 0
π 0
Join us on May 13 for the Inno4scale Innovation Day in Barcelona π‘π
This event marks the successful conclusion of the Inno4scale project and provides a platform for future collaborations in #HPC π₯οΈπ©βπ¬
Registrationπ regi.hlrs.de/2025/inno4sc...
We look forward to meet you in Barcelona! β¨
15.04.2025 09:02 β
π 0
π 1
π¬ 0
π 1
Simulating space plasma at ion-kinetic accuracy demands exascale computing! But checkpointing is a challenge. Discuss in the comments:
π¬ What is AIβs advantage in checkpointing?
1οΈβ£ Less simulation noise
2οΈβ£ Higher plasma density
3οΈβ£ Faster storage & recovery
Look for hints ππ www.inno4scale.eu
09.04.2025 09:00 β
π 0
π 0
π¬ 0
π 0
inno4scale
Inno4scale is a European initiative, which was started to support the development of innovative algorithms for exascale supercomputers, so their efficient use can be fully exploited. Currentlyβ¦
GW calculations reveal materials insights, while scaling remains diffucult. Discuss in the comments on the following question π‘π§
π¬ What is the main bottleneck in large-scale GW?
1οΈβ£ Polarizability calculation
2οΈβ£ Wavefunction storage
3οΈβ£ Screened Coulomb W
Look for hints ππ www.inno4scale.eu
03.04.2025 06:30 β
π 0
π 0
π¬ 0
π 0
Join us on May 13 for the Inno4scale Innovation Day in Barcelona π‘π
This event marks the successful conclusion of the Inno4scale project and provides a platform for future collaborations in #HPC π₯οΈπ©βπ¬
Registrationπ regi.hlrs.de/2025/inno4sc...
We look forward to meet you in Barcelona! β¨
01.04.2025 09:00 β
π 2
π 0
π¬ 0
π 1
inno4scale
Inno4scale is a European initiative, which was started to support the development of innovative algorithms for exascale supercomputers, so their efficient use can be fully exploited. Currently existing codes for high-performance computing will not be able to function efficiently on upcoming exascale...
HPC advances simulations but communication overhead limits scalability. Discuss in the comments on the following question π‘π§
π¬ What improves CG solver efficiency on GPUs?
1οΈβ£ Kernel launch overhead
2οΈβ£ Floating-point precision
3οΈβ£ Memory bandwidth
More information and hints βΆοΈ
26.03.2025 07:31 β
π 1
π 0
π¬ 0
π 0
inno4scale
Inno4scale is a European initiative, which was started to support the development of innovative algorithms for exascale supercomputers, so their efficient use can be fully exploited. Currently existing codes for high-performance computing will not be able to function efficiently on upcoming exascale...
Lattice QCD pushes precision physics, but scaling remains difficult. Discuss in the comments on the following question π‘π§
π¬What is a major challenge in Lattice QCD?
1οΈβ£ Critical slowing down
2οΈβ£ Low lattice resolution
3οΈβ£ No experimental proof
More insights and hints βΆοΈ
20.03.2025 07:31 β
π 1
π 0
π¬ 0
π 0
Join us on May 13 for the Inno4scale Innovation Day in Barcelona π‘πΒ
This event marks the successful conclusion of the Inno4scale project and provides a platform for future collaborations in #HPC π₯οΈπ©βπ¬Β
Registrationπ https://regi.hlrs.de/2025/inno4scale/registration/registrationForm.jsp
18.03.2025 10:01 β
π 2
π 2
π¬ 0
π 0
inno4scale
Inno4scale is a European initiative, which was started to support the development of innovative algorithms for exascale supercomputers, so their efficient use can be fully exploited. Currently existing codes for high-performance computing will not be able to function efficiently on upcoming exascale...
HPC accelerates simulations, but scaling remains a challenge. Discuss in the comments on the following questionπ‘π§
π¬ How do Neural Operators improve PinT?
1οΈβ£ No need for fine solvers
2οΈβ£ Faster with high accuracy
3οΈβ£ Replaces all numerical models
More information and hints βΆοΈ
12.03.2025 07:31 β
π 0
π 0
π¬ 0
π 0
inno4scale
Inno4scale is a European initiative, which was started to support the development of innovative algorithms for exascale supercomputers, so their efficient use can be fully exploited. Currently existing codes for high-performance computing will not be able to function efficiently on upcoming exascale...
HPC drives CFD forward, but scalability remains a challenge.Β Discuss in the comments on the following question π‘π§
π¬ How does ML improve CFD scalability?
1. Replaces Navier-Stokes
2. Speeds up solvers & coupling
3. Removes computational grids
More information and hints βΆοΈ
05.03.2025 07:31 β
π 0
π 0
π¬ 0
π 0
Now, you have had the chance to learn a lot about our innovation studies π§ Β Β
Over the next few weeks, you can look forward to some exciting quizzes on our innovation studies and related topics! π₯οΈ π’Β
You are invited to visit our website to refresh your knowledge βΆοΈ https://www.inno4scale.eu/
26.02.2025 07:30 β
π 1
π 0
π¬ 0
π 0