26 citations · 63 across the 14 of their papers we have counts for
17 papers
Memory-Efficient Expert Routing for Distributed MoE Training
Arnab Kanti Tarafder, Jaume Guasch-Martí, Gokcen Kestor +1
As Mixture-of-Experts (MoE) models scale toward hundreds of experts and higher top- routing, memory efficiency in distributed training becomes a critical bottleneck. Peak memory…
GroupMask: Layer-Adaptive Group-wise Sparsity for Semi-Structured LLM Pruning
Zhengao Li, Shuoqiu Li, Xiaofang Zhang +7
Semi-structured pruning compresses large language models (LLMs) while keeping a regular sparse structure, but the prevailing N:M pattern fixes the same local sparsity ratio in ever…
RuC: HDL-Agnostic Rule Completion Benchmark Generation
Arnau Ayguadé Domingo, Miquel Alberti-Binimelis, Cristian Gutierrez-Gomez +5
Large Language Models (LLMs) have rapidly improved in performance across code-related tasks, making their integration into Register Transfer Level (RTL) development increasingly at…
NotSoTiny: A Large, Living Benchmark for RTL Code Generation
Razine Moundir Ghorab, Emanuele Parisi, Cristian Gutierrez +4
LLMs have shown early promise in generating RTL code, yet evaluating their capabilities in realistic setups remains a challenge. So far, RTL benchmarks have been limited in scale,…
TuRTLe: A Unified Evaluation of LLMs for RTL Generation
Dario Garcia-Gasulla, Gokcen Kestor, Emanuele Parisi +6
The rapid advancements in LLMs have driven the adoption of generative AI in various domains, including Electronic Design Automation (EDA). Unlike traditional software development,…
Union: A Unified HW-SW Co-Design Ecosystem in MLIR for Evaluating Tensor Operations on Spatial Accelerators
Geonhwa Jeong, Gokcen Kestor, Prasanth Chatarasi +5
To meet the extreme compute demands for deep learning across commercial and scientific applications, dataflow accelerators are becoming increasingly popular. While these "domain-sp…
COMET: A Domain-Specific Compilation of High-Performance Computational Chemistry
Erdal Mutlu, Ruiqin Tian, Bin Ren +4
The computational power increases over the past decades havegreatly enhanced the ability to simulate chemical reactions andunderstand ever more complex transformations. Tensor cont…
A High-Performance Sparse Tensor Algebra Compiler in Multi-Level IR
Ruiqin Tian, Luanzheng Guo, Jiajia Li +2
Tensor algebra is widely used in many applications, such as scientific computing, machine learning, and data analytics. The tensors represented real-world data are usually large an…
Smart-PGSim: Using Neural Network to Accelerate AC-OPF Power Grid Simulation
Wenqian Dong, Zhen Xie, Gokcen Kestor +1
The optimal power flow (OPF) problem is one of the most important optimization problems for the operation of the power grid. It calculates the optimum scheduling of the committed g…
MPI Windows on Storage for HPC Applications
Sergio Rivas-Gomez, Roberto Gioiosa, Ivy Bo Peng +4
Upcoming HPC clusters will feature hybrid memories and storage devices per compute node. In this work, we propose to use the MPI one-sided communication model and MPI windows as un…
Runtime Concurrency Control and Operation Scheduling for High Performance Neural Network Training
Jiawen Liu, Dong Li, Gokcen Kestor +1
Training neural network often uses a machine learning framework such as TensorFlow and Caffe2. These frameworks employ a dataflow model where the NN training is modeled as a direct…
MPI Streams for HPC Applications
Ivy Bo Peng, Stefano Markidis, Roberto Gioiosa +2
Data streams are a sequence of data flowing between source and destination processes. Streaming is widely used for signal, image and video processing for its efficiency in pipelini…
Preparing HPC Applications for the Exascale Era: A Decoupling Strategy
Ivy Bo Peng, Roberto Gioiosa, Gokcen Kestor +2
Production-quality parallel applications are often a mixture of diverse operations, such as computation- and communication-intensive, regular and irregular, tightly coupled and loo…
Extending Message Passing Interface Windows to Storage
Sergio Rivas-Gomez, Stefano Markidis, Ivy Bo Peng +3
This work presents an extension to MPI supporting the one-sided communication model and window allocations in storage. Our design transparently integrates with the current MPI impl…
Exploring the Performance Benefit of Hybrid Memory System on HPC Environments
Ivy Bo Peng, Roberto Gioiosa, Gokcen Kestor +2
Hardware accelerators have become a de-facto standard to achieve high performance on current supercomputers and there are indications that this trend will increase in the future. M…
Idle Period Propagation in Message-Passing Applications
Ivy Bo Peng, Stefano Markidis, Erwin Laure +2
Idle periods on different processes of Message Passing applications are unavoidable. While the origin of idle periods on a single process is well understood as the effect of system…
Exploring Application Performance on Emerging Hybrid-Memory Supercomputers
Ivy Bo Peng, Stefano Markidis, Erwin Laure +2
Next-generation supercomputers will feature more hierarchical and heterogeneous memory systems with different memory technologies working side-by-side. A critical question is wheth…