Scalable Parallel Computing Lab, SPCL @ ETH Zurich

@UC6Yn2CBgLJeW60chSkWWC5g - 4.4K subscribers

Talks from SPCL@ETH members and SPCL_Bcast episodes!

Home Videos Live Playlists

quatrex: Quantum Transport at the Exascale

quatrex: Quantum Transport at the Exascale Scalable Parallel Computing Lab, SPCL @ ETH Zurich

70 views - 5 days ago

Keynote at SCA26 - Torsten Hoefler: Ultra Ethernet for next generation AI and HPC workloads

Keynote at SCA26 - Torsten Hoefler: Ultra Ethernet for next generation AI and HPC workloads Scalable Parallel Computing Lab, SPCL @ ETH Zurich

154 views - 1 month ago

ATLAHS: An Application-centric Network Simulator Toolchain for AI, HPC, and Distributed Storage

ATLAHS: An Application-centric Network Simulator Toolchain for AI, HPC, and Distributed Storage Scalable Parallel Computing Lab, SPCL @ ETH Zurich

85 views - 1 month ago

A Conversation with Afif Boudaoud about Automatic Differentiation and DaCe

A Conversation with Afif Boudaoud about Automatic Differentiation and DaCe Scalable Parallel Computing Lab, SPCL @ ETH Zurich

237 views - 3 months ago

C.A.T.S. - Improved Whole Program Optimization

C.A.T.S. - Improved Whole Program Optimization Scalable Parallel Computing Lab, SPCL @ ETH Zurich

291 views - 3 months ago

Computing the Full Earth System at 1km Resolution

Computing the Full Earth System at 1km Resolution Scalable Parallel Computing Lab, SPCL @ ETH Zurich

157 views - 5 months ago

Uno: A One-Stop Solution for Inter- & Intra-Data Center Congestion Control and Reliable Connectivity

Uno: A One-Stop Solution for Inter- & Intra-Data Center Congestion Control and Reliable Connectivity Scalable Parallel Computing Lab, SPCL @ ETH Zurich

113 views - 5 months ago

PerfDojo: Automated ML Library Generation for Heterogeneous Architectures

PerfDojo: Automated ML Library Generation for Heterogeneous Architectures Scalable Parallel Computing Lab, SPCL @ ETH Zurich

113 views - 7 months ago

Can we build an AI Climate Scientist?

Can we build an AI Climate Scientist? Scalable Parallel Computing Lab, SPCL @ ETH Zurich

241 views - 7 months ago

Reproducibility of AI Models (A Tiny Note)

Reproducibility of AI Models (A Tiny Note) Scalable Parallel Computing Lab, SPCL @ ETH Zurich

344 views - 10 months ago

FPsPIN: An FPGA-based Open-Hardware Research Platform for Processing in the Network

FPsPIN: An FPGA-based Open-Hardware Research Platform for Processing in the Network Scalable Parallel Computing Lab, SPCL @ ETH Zurich

168 views - 10 months ago

ARMing GPUs: On the Memory Subsystem of Grace Hopper GH200

ARMing GPUs: On the Memory Subsystem of Grace Hopper GH200 Scalable Parallel Computing Lab, SPCL @ ETH Zurich

934 views - 10 months ago

What is AI for Science - Models, Data, and Computation

What is AI for Science - Models, Data, and Computation Scalable Parallel Computing Lab, SPCL @ ETH Zurich

544 views - 11 months ago

Reasoning Language Models Will Solve All Our Problems (given the right machines)

Reasoning Language Models Will Solve All Our Problems (given the right machines) Scalable Parallel Computing Lab, SPCL @ ETH Zurich

686 views - 11 months ago

LLAMP: Assessing Network Latency Sensitivity Tolerance of HPC Applications with Linear Programming

LLAMP: Assessing Network Latency Sensitivity Tolerance of HPC Applications with Linear Programming Scalable Parallel Computing Lab, SPCL @ ETH Zurich

149 views - 1 year ago

Chameleon: Heterogeneous & Disaggregated Accelerator System for Retrieval-Augmented Language Models

Chameleon: Heterogeneous & Disaggregated Accelerator System for Retrieval-Augmented Language Models Scalable Parallel Computing Lab, SPCL @ ETH Zurich

140 views - 1 year ago

Offloaded MPI Message Matching: An Optimistic Approach

Offloaded MPI Message Matching: An Optimistic Approach Scalable Parallel Computing Lab, SPCL @ ETH Zurich

161 views - 1 year ago

EDAN: Towards Understanding Memory Parallelism and Latency Sensitivity in HPC

EDAN: Towards Understanding Memory Parallelism and Latency Sensitivity in HPC Scalable Parallel Computing Lab, SPCL @ ETH Zurich

167 views - 1 year ago

ARMing GPUs for Impactful Science with the GH200 Superchip (and true Exascale)

ARMing GPUs for Impactful Science with the GH200 Superchip (and true Exascale) Scalable Parallel Computing Lab, SPCL @ ETH Zurich

1.2K views - 1 year ago

RapidChiplet: How to Explore the Design Space of Inter-Chiplet Interconnects

RapidChiplet: How to Explore the Design Space of Inter-Chiplet Interconnects Scalable Parallel Computing Lab, SPCL @ ETH Zurich

405 views - 1 year ago

MARLIN: Mixed-Precision Auto-Regressive Parallel Inference on Large Language Models

MARLIN: Mixed-Precision Auto-Regressive Parallel Inference on Large Language Models Scalable Parallel Computing Lab, SPCL @ ETH Zurich

236 views - 1 year ago

Proximity is all You Need– Two Tricks for Chiplet Interconnects

Proximity is all You Need– Two Tricks for Chiplet Interconnects Scalable Parallel Computing Lab, SPCL @ ETH Zurich

768 views - 1 year ago

Exploring GPU-to-GPU Communication: Insights into Supercomputer Interconnects

Exploring GPU-to-GPU Communication: Insights into Supercomputer Interconnects Scalable Parallel Computing Lab, SPCL @ ETH Zurich

450 views - 1 year ago

[SPCL_Bcast] Measurement and Analysis of Application Performance on Exascale GPU-accelerated Systems

[SPCL_Bcast] Measurement and Analysis of Application Performance on Exascale GPU-accelerated Systems Scalable Parallel Computing Lab, SPCL @ ETH Zurich

233 views - 1 year ago

[SPCL_Bcast] Data Selection - Data Challenges when Training Generative Models

[SPCL_Bcast] Data Selection - Data Challenges when Training Generative Models Scalable Parallel Computing Lab, SPCL @ ETH Zurich

139 views - 1 year ago

All models are wrong, some are useful: Model Selection with Limited Labels

All models are wrong, some are useful: Model Selection with Limited Labels Scalable Parallel Computing Lab, SPCL @ ETH Zurich

122 views - 1 year ago

High Performance Unstructured SpMM Computation Using Tensor Cores

High Performance Unstructured SpMM Computation Using Tensor Cores Scalable Parallel Computing Lab, SPCL @ ETH Zurich

203 views - 1 year ago

[SPCL_Bcast] Merging and MoErging for compositional generalization

[SPCL_Bcast] Merging and MoErging for compositional generalization Scalable Parallel Computing Lab, SPCL @ ETH Zurich

498 views - 1 year ago

Iterating Pointers: Enabling Static Analysis for Loop-based Pointers

Iterating Pointers: Enabling Static Analysis for Loop-based Pointers Scalable Parallel Computing Lab, SPCL @ ETH Zurich

226 views - 1 year ago

[SPCL_Bcast #54] Broadcast, Reduction and beyond with Block Schedules and Circulant Graphs

[SPCL_Bcast #54] Broadcast, Reduction and beyond with Block Schedules and Circulant Graphs Scalable Parallel Computing Lab, SPCL @ ETH Zurich

215 views - 1 year ago

[SPCL_Bcast #53] The evolution of accelerator-centric GPU services - past, present, future

[SPCL_Bcast #53] The evolution of accelerator-centric GPU services - past, present, future Scalable Parallel Computing Lab, SPCL @ ETH Zurich

295 views - 1 year ago

[SPCL_Bcast #52] Deep learning can beat numerical weather prediction! What's next?

[SPCL_Bcast #52] Deep learning can beat numerical weather prediction! What's next? Scalable Parallel Computing Lab, SPCL @ ETH Zurich

253 views - 1 year ago

From Large Language Models to Reasoning Language Models - Three Eras in The Age of Computation.

From Large Language Models to Reasoning Language Models - Three Eras in The Age of Computation. Scalable Parallel Computing Lab, SPCL @ ETH Zurich

14K views - 1 year ago

[SPCL_Bcast #51] Neural Network Quantization with Brevitas

[SPCL_Bcast #51] Neural Network Quantization with Brevitas Scalable Parallel Computing Lab, SPCL @ ETH Zurich

855 views - 1 year ago

How to Leverage Dataflow To Find and Squash Program Optimization Bugs

How to Leverage Dataflow To Find and Squash Program Optimization Bugs Scalable Parallel Computing Lab, SPCL @ ETH Zurich

171 views - 1 year ago

[SPCL_Bcast #48] Improving Cloud Security with Hardware Memory Capabilities

[SPCL_Bcast #48] Improving Cloud Security with Hardware Memory Capabilities Scalable Parallel Computing Lab, SPCL @ ETH Zurich

113 views - 1 year ago

[SPCL_Bcast #49] Programming Groq LPUs without IEEE Floating Point

[SPCL_Bcast #49] Programming Groq LPUs without IEEE Floating Point Scalable Parallel Computing Lab, SPCL @ ETH Zurich

163 views - 1 year ago

[SPCL_Bcast #50] Hardware-aware Algorithms for Language Modeling

[SPCL_Bcast #50] Hardware-aware Algorithms for Language Modeling Scalable Parallel Computing Lab, SPCL @ ETH Zurich

510 views - 1 year ago

How to find Relevant Items using Approximate Nearest Neighbor Search

How to find Relevant Items using Approximate Nearest Neighbor Search Scalable Parallel Computing Lab, SPCL @ ETH Zurich

297 views - 1 year ago

Exascale Cloud Computing – A Foggy Tale of Networks, AI, Containers, and Ultra Ethernet

Exascale Cloud Computing – A Foggy Tale of Networks, AI, Containers, and Ultra Ethernet Scalable Parallel Computing Lab, SPCL @ ETH Zurich

525 views - 1 year ago

Swing: Short-cutting Rings for Higher Bandwidth Allreduce

Swing: Short-cutting Rings for Higher Bandwidth Allreduce Scalable Parallel Computing Lab, SPCL @ ETH Zurich

263 views - 2 years ago

Neural Graph Databases

Neural Graph Databases Scalable Parallel Computing Lab, SPCL @ ETH Zurich

271 views - 2 years ago

HOT - Higher-Order Dynamic Graph Representation Learning with Efficient Transformers

HOT - Higher-Order Dynamic Graph Representation Learning with Efficient Transformers Scalable Parallel Computing Lab, SPCL @ ETH Zurich

243 views - 2 years ago

LRSCwait: Enabling Scalable and Efficient Synchronization in Manycore Systems

LRSCwait: Enabling Scalable and Efficient Synchronization in Manycore Systems Scalable Parallel Computing Lab, SPCL @ ETH Zurich

134 views - 2 years ago

Compressing Multidimensional Weather and Climate Data Into Neural Networks

Compressing Multidimensional Weather and Climate Data Into Neural Networks Scalable Parallel Computing Lab, SPCL @ ETH Zurich

271 views - 2 years ago

VENOM: A Vectorized N:M Format for Unleashing the Power of Sparse Tensor Cores

VENOM: A Vectorized N:M Format for Unleashing the Power of Sparse Tensor Cores Scalable Parallel Computing Lab, SPCL @ ETH Zurich

743 views - 2 years ago

Motif Prediction with Graph Neural Networks

Motif Prediction with Graph Neural Networks Scalable Parallel Computing Lab, SPCL @ ETH Zurich

565 views - 2 years ago

Demystifying Chains, Trees, and Graphs of Thoughts

Demystifying Chains, Trees, and Graphs of Thoughts Scalable Parallel Computing Lab, SPCL @ ETH Zurich

929 views - 2 years ago

[SPCL_Bcast #47] The digital revolution of Earth system modelling

[SPCL_Bcast #47] The digital revolution of Earth system modelling Scalable Parallel Computing Lab, SPCL @ ETH Zurich

217 views - 2 years ago

[SPCL_Bcast #46] Capturing Computation with Algorithmic Alignment

[SPCL_Bcast #46] Capturing Computation with Algorithmic Alignment Scalable Parallel Computing Lab, SPCL @ ETH Zurich

233 views - 2 years ago

Co-design Hardware and Algorithm for Vector Search

Co-design Hardware and Algorithm for Vector Search Scalable Parallel Computing Lab, SPCL @ ETH Zurich

359 views - 2 years ago

Demystifying Graph Databases

Demystifying Graph Databases Scalable Parallel Computing Lab, SPCL @ ETH Zurich

205 views - 2 years ago

Fortran is dead – Long live Fortran!

Fortran is dead – Long live Fortran! Scalable Parallel Computing Lab, SPCL @ ETH Zurich

3.7K views - 2 years ago

Hot Interconnects - EtherNET: the present and future of datacenter and supercomputers

Hot Interconnects - EtherNET: the present and future of datacenter and supercomputers Scalable Parallel Computing Lab, SPCL @ ETH Zurich

420 views - 2 years ago

[SPCL_Bcast] Can I Cook a 5 o'clock Compiler Cake and Eat It at 2?

[SPCL_Bcast] Can I Cook a 5 o'clock Compiler Cake and Eat It at 2? Scalable Parallel Computing Lab, SPCL @ ETH Zurich

235 views - 2 years ago

AI-Driven Performance Metaprogramming

AI-Driven Performance Metaprogramming Scalable Parallel Computing Lab, SPCL @ ETH Zurich

601 views - 2 years ago

HammingMesh: A Network Topology for Large-Scale Deep Learning

HammingMesh: A Network Topology for Large-Scale Deep Learning Scalable Parallel Computing Lab, SPCL @ ETH Zurich

947 views - 2 years ago

GDI: Scaling Online Transactional and Analytical Graph Workloads to Hundreds of Thousands of Cores

GDI: Scaling Online Transactional and Analytical Graph Workloads to Hundreds of Thousands of Cores Scalable Parallel Computing Lab, SPCL @ ETH Zurich

141 views - 2 years ago

[SPCL_Bcast] Scalable Graph Machine Learning

[SPCL_Bcast] Scalable Graph Machine Learning Scalable Parallel Computing Lab, SPCL @ ETH Zurich

198 views - 2 years ago

[SPCL_Bcast] Heterogeneous multi-core systems for efficient EdgeML

[SPCL_Bcast] Heterogeneous multi-core systems for efficient EdgeML Scalable Parallel Computing Lab, SPCL @ ETH Zurich

423 views - 2 years ago