Marcin Copik

From Fortran to data-centric programming to accelerate climate simulations - our team has won this year's ACM Gordon Bell Prize for Climate Modeling!

Congratulations to our entire team - Alexandru, Yakup, Pratyai, Ben, and @thoefler.bsky.social!

#SC25 #HPC #HPCignites

GitHub - spcl/xaas-containers Contribute to spcl/xaas-containers development by creating an account on GitHub.

⚙️ Code: github.com/spcl/xaas-containers
📚 Artifact: zenodo.org/records/1711...

Sadly, I won't be able to attend #SC25 due to medical reasons. Nonetheless, I'm looking forward to @thoefler.bsky.social presenting our performance-portable XaaS containers! If you're interested in #HPC containers, then join us on Tuesday, 3:30pm, Room 275 - eager to hear your thoughts! #HPCignites

XaaS Containers: Performance-Portable Representation With Source and IR Containers | Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis

Join us on Tuesday in room 275!

📄Paper: dl.acm.org/doi/10.1145/...
💬Presentation: Tuesday, November 18th, 3:30pm - 3:52pm CST, Room 275

XaaS containers are the result of a collaboration between SPCL, @cscsch.bsky.social, LLNL, ANL, and the University of Chicago.

#HPC #SC25 #HPCignites

We designed Source and Intermediate Representation (IR) containers. In IR containers, we implement a pipeline where containerized applications are compiled to LLVM IR, and adapted to the target architecture (e.g., vectorization) at deployment time, ensuring peak performance.

Traditional containers lock in compilation decisions too early, hurting performance on diverse HPC systems. We delay performance-critical decisions until the target system is known, avoiding the traditional container trade-off between portability and hardware-specific optimizations.

This year at #SC25, we will present XaaS containers, the Best Paper Finalist at 🏆! We propose a new paradigm for performance portability to bridge the gap between #cloud productivity and #HPC performance 🚀

#HPCignites @thoefler.bsky.social

Cppless: Single-Source and High-Performance Serverless Programming in C++ The rise of serverless computing introduced a new class of scalable, elastic and widely available parallel workers in the cloud. Many systems and applications benefit from offloading computations and ...

Cppless is open source and built on top of LLVM with less than 1k LoC changes. More details in the paper on serialization, C++ lambda extraction, and cross-compilation.

Paper 👉 arxiv.org/abs/2401.10834
Code 👉 github.com/spcl/cppless
Artifact 👉 zenodo.org/records/1577...

Our evaluation shows that C++ serverless functions can scale to 512 parallel workers with a double-digit millisecond overhead. Using the example of ray tracing, we demonstrate a speedup of up to 59x. From a 60-second execution time to 1 second with hundreds of functions and a minimal cost increase.

With Cppless, C++ is no longer a second-class citizen in serverless. We introduce alternative entry points and implement lambda reflection, making C++ functions as easy to deploy as Python.

Cold starts? Just 10-12 milliseconds and up to 60x faster than Python functions!

🚀Cppless enables single-source #serverless programming in C++! No more splitting functions from the main code - just add a few lines to accelerate your program by offloading to systems like AWS Lambda. Our paper on Clang-based #FaaS for C++ has just been accepted at ACM TACO.

⭐Larissa Schmid will present at EuroSys her paper "SeBS-Flow: Benchmarking Serverless Cloud Function Workflows".
📅 Thursday, April 3, 2025 at 9.00. Session 7-1
📚 Paper: arxiv.org/abs/2410.03480

⭐At the EuroSys poster session, I will present our joint work on SeBS-Flow, a new extension of SeBS that provides automatic and cloud-agnostic benchmarking of serverless workflows.
📅 Monday, March 31, 2025 at 18:00 - 19:30

⭐ I will present a work-in-progress paper at the SESAME'25 workshop: "SeBS 2.0: Keeping up with the Clouds"
📅 Monday, March 31, 2025 at 11:00 - 12:10
📚 Paper: dl.acm.org/doi/10.1145/...

I will be attending EuroSys/ASPLOS 2025, where I will present our work on next-generation #serverless benchmarking. If you're going to be there and would like to talk about serverless, please feel free to reach out!

This year, we present two papers and a poster!

Reasoning Language Models: A Blueprint

📜 arxiv.org/abs/2501.11223
x1 on GitHub: github.com/spcl/x1

We introduce x1, a modular implementation for rapid RLM prototyping, provide key insights (multi-phase training for policy/value models, importance of familiar training distributions) and discuss scalable RLM cloud deployments and how RLMs fit into the broader LLM ecosystem.

We propose a blueprint that organizes RLM components into a modular framework, based on an analysis of all RLM works, which incorporates reasoning structures (chains, trees, graphs) and strategies (e.g., #MCTS), Reinforcement Learning and supervision schemes.

Reasoning language models (#RLMs), such as o1 or QwQ, have redefined AI's problem-solving capabilities by extending #LLMs with advanced reasoning mechanisms. Yet, their high costs, proprietary nature, and complex architectures present accessibility and scalability challenges.

Congratulations to Marcin Kopic and @thoefler.bsky.social @SPCL ETH Zurich for #SC24 Best Research Poster Award!

Kyle Chard opens panelists statements at #SC24 panel on #serverless #HPC. Join us in room B314!

The #SC24 panel on #serverless #HPC happens today at 10:30 am! Looking forward to the discussion on the future Cloud & HPC with our panelists: @ianfoster42.bsky.social, @thoefler.bsky.social, Satoshi Matsuoka, Kyle Chard, and Devesh Tiwari!

Fun times at the #SC24 CANOPIE workshop panel on aXeleration as a Service (XaaS) at #SC24. Nice discussion with @ernstdj.bsky.social, Ian Foster, and Thomas Schulthess!

Nicely moderated by @mcopik.bsky.social.

Great attendance at the #SC24 panel on performance portable containers #XaaS! Interesting conclusions on the future of #HPC and #cloud, in a discussion between @danernst.bsky.social, @ianfoster42.bsky.social, Torsten Hoefler, and Thomas Schulthess!

Interesting observation, I hope that you will join the discussion on Thursday :)

Happy to meet&chat at SC to discuss your experience and thoughts!

I have the pleasure of moderating two panels this year at #SC24, and the second one focuses on the core part of my PhD thesis: #serverless #HPC.

Join us next Thursday and participate in the discussion on the future of #cloud programming models #supercomputing!

The panel brings experts from academia and national labs: Kyle Chard, Ian Foster, Torsten Hoefler, Satoshi Matsuoka, and Devesh Tiwari.

📅 Thursday, 21st November
⏰ 10:30 AM EST
🏢 Room B313B-314

Is #serverless a fad that will disappear in a few years, or will it accelerate #HPC - #Cloud convergence and change the #Supercomputing landscape?

Join us at the #SC24 panel "Serverless HPC: Challenges, Opportunities, and Future Prospects for Accelerated Cloud Computing"!

The panel includes authors of the XaaS vision: Dan Ernst, Ian Foster, Torsten Hoefler, and Thomas Schulthess!

XaaS paper 👉 spcl.inf.ethz.ch/Publications...

Event page 👉 sc24.conference-program.com/presentation...

Can high-performance containers offer performance portability and unified abstraction for the converged #HPC and #Cloud? Join us at the #SC24 CANOPIE-HPC workshop for a panel on the changing HPC landscape.

📅 Sunday, 17th November
⏰ 11:40 AM EST
🏢 #Supercomputing Room B313