Related papers: ChipletPart: Cost-Aware Partitioning for 2.5D Syst…
Large Language Models (LLMs) have achieved remarkable success in various fields, but their training and finetuning require massive computation and memory, necessitating parallelism which introduces heavy communication overheads. Driven by…
Multi-Chip-Modules (MCMs) reduce the design and fabrication cost of machine learning (ML) accelerators while delivering performance and energy efficiency on par with a monolithic large chip. However, ML compilers targeting MCMs need to…
The increasing complexity and cost of manufacturing monolithic chips have driven the semiconductor industry toward chiplet-based designs, where smaller and modular chiplets are integrated onto a single interposer. While chiplet…
The advent of chiplet technology introduces cutting-edge opportunities for constructing highly heterogeneous platforms with specialized accelerators. However, the HPC community currently lacks expertise in hardware development, a gap that…
Leveraging 2.5D interposer technology, we advocate the integration of untrusted commodity components/chiplets with physically separate, entrusted logic components. Such organization provides a modern root of trust for secure system-level…
Current algorithms for large-scale industrial optimization problems typically face a trade-off: they either require exponential time to reach optimal solutions, or employ problem-specific heuristics. To overcome these limitations, we…
Hypergraph partitioning is an important problem in machine learning, computer vision and network analytics. A widely used method for hypergraph partitioning relies on minimizing a normalized sum of the costs of partitioning hyperedges…
Modern Artificial Intelligence (AI) workloads demand computing systems with large silicon area to sustain throughput and competitive performance. However, prohibitive manufacturing costs and yield limitations at advanced tech nodes and…
As program workloads (e.g., AI) increase in size and algorithmic complexity, the primary challenge lies in their high dimensionality, encompassing computing cores, array sizes, and memory hierarchies. To overcome these obstacles, innovative…
Modern AI acceleration faces a fundamental challenge: conventional assumptions about memory requirements, batching effectiveness, and latency-throughput tradeoffs are systemwide generalizations that ignore the heterogeneous computational…
Floorplanning for systems-on-a-chip (SoCs) and its sub-systems is a crucial and non-trivial step of the physical design flow. It represents a difficult combinatorial optimization problem. A typical large scale SoC with 120 partitions…
This paper presents the HiPart package, an open-source native python library that provides efficient and interpret-able implementations of divisive hierarchical clustering algorithms. HiPart supports interactive visualizations for the…
The emergence of chiplet-based heterogeneous integration is transforming the semiconductor, AI, and high-performance computing industries by enabling modular designs and improved scalability. However, assembling chiplets from multiple…
Transformers have revolutionized deep learning and generative modeling, enabling advancements in natural language processing tasks. However, the size of transformer models is increasing continuously, driven by enhanced capabilities across…
When partitioning gate-level netlists using graphs, it is beneficial to cluster gates to reduce the order of the graph and preserve some characteristics of the circuit that the partitioning might degrade. Gate clustering is even more…
Balanced partitioning is often a crucial first step in solving large-scale graph optimization problems, e.g., in some cases, a big graph can be chopped into pieces that fit on one machine to be processed independently before stitching the…
Fully homomorphic encryption (FHE) is in the spotlight as a definitive solution for privacy, but the high computational overhead of FHE poses a challenge to its practical adoption. Although prior studies have attempted to design ASIC…
Embedding learning is an important technique in deep recommendation models to map categorical features to dense vectors. However, the embedding tables often demand an extremely large number of parameters, which become the storage and…
Partitioning a graph into blocks of roughly equal weight while cutting only few edges is a fundamental problem in computer science with numerous practical applications. While shared-memory parallel partitioners have recently matured to…
To address increasing compute demand from recent multi-model workloads with heavy models like large language models, we propose to deploy heterogeneous chiplet-based multi-chip module (MCM)-based accelerators. We develop an advanced…