Related papers: Hive Hash Table: A Warp-Cooperative, Dynamically R…
Industrial systems increasingly depend on Machine Learning (ML), and operate on heterogeneous nodes that must satisfy tight latency, energy, and memory constraints. Dynamic ML models, which reconfigure their computational footprint at…
Deep Learning system architects strive to design a balanced system where the computational accelerator -- FPGA, GPU, etc, is not starved for data. Feeding training data fast enough to effectively keep the accelerator utilization high is…
Processing long-context inputs with large language models presents a significant challenge due to the enormous memory requirements of the Key-Value (KV) cache during inference. Existing KV cache compression methods exhibit noticeable…
Die-stacked DRAM has been proposed for use as a large, high-bandwidth, last-level cache with hundreds or thousands of megabytes of capacity. Not all workloads (or phases) can productively utilize this much cache space, however.…
The conventional model of resource allocation in HPC systems is static. Thus, a job cannot leverage newly available resources in the system or release underutilized resources during the execution. In this paper, we present Kub, a…
Retrieval-Augmented Generation (RAG) encounters efficiency challenges when scaling to massive knowledge bases while preserving contextual relevance. We propose Hash-RAG, a framework that integrates deep hashing techniques with systematic…
The significant resource demands in LLM serving prompts production clusters to fully utilize heterogeneous hardware by partitioning LLM models across a mix of high-end and low-end GPUs. However, existing parallelization approaches often…
Retrieval-Augmented Generation (RAG) systems suffer from severe time-to-first-token (TTFT) bottlenecks due to long input sequences. Existing KV cache reuse methods face a fundamental trade-off: prefix caching requires identical prefixes…
Emerging reconfigurable datacenters allow to dynamically adjust the network topology in a demand-aware manner. These datacenters rely on optical switches which can be reconfigured to provide direct connectivity between racks, in the form of…
Hierarchical matrices are data-sparse approximations of dense matrices that are widely used for fast matrix computations. Hierarchical matrices are built using a tree data structure, with low-rank blocks identified by various admissibility…
Connected components and spanning forest are fundamental graph algorithms due to their use in many important applications, such as graph clustering and image segmentation. GPUs are an ideal platform for graph algorithms due to their high…
Owing to the significance of combinatorial search strategies both for academia and industry, the introduction of new techniques is a fast growing research field these days. These strategies have really taken different forms ranging from…
The growing data has brought tremendous pressure for query processing and storage, so there are many studies that focus on using GPU to accelerate join operation, which is one of the most important operations in modern database systems.…
Current approaches to scheduling workloads on heterogeneous systems with specialized accelerators often rely on manual partitioning, offloading tasks with specific compute patterns to accelerators. This method requires extensive…
Processing-in-cache (PiC) and Processing-in-memory (PiM) architectures, especially those utilizing bit-line computing, offer promising solutions to mitigate data movement bottlenecks within the memory hierarchy. While previous studies have…
Persistent memory is the bottleneck separating stateless chatbots from long-running agentic systems. Retrieval-augmented generation (RAG) over flat vector stores fragments facts into chunks, loses cross-session identity, and has no…
Dynamic graphs, featuring continuously updated vertices and edges, have grown in importance for numerous real-world applications. To accommodate this, graph frameworks, particularly their internal data structures, must support both…
Carefully balancing load in distributed stream processing systems has a fundamental impact on execution latency and throughput. Load balancing is challenging because real-world workloads are skewed: some tuples in the stream are associated…
In industrial recommendation systems on websites and apps, it is essential to recall and predict top-n results relevant to user interests from a content pool of billions within milliseconds. To cope with continuous data growth and improve…
Log parsing, which extracts log templates from semi-structured logs and produces structured logs, is the first and the most critical step in automated log analysis. While existing log parsers have achieved decent results, they suffer from…