Related papers: Mosaic: An Application-Transparent Hardware-Softwa…
In cloud machine learning (ML) inference systems, providing low latency to end-users is of utmost importance. However, maximizing server utilization and system throughput is also crucial for ML service providers as it helps lower the…
Online analytical processing of queries on datasets in the many-terabyte range is only possible with costly distributed computing systems. To decrease the cost and increase the throughput, systems can leverage accelerators such as GPUs,…
Massively multicore processors, such as Graphics Processing Units (GPUs), provide, at a comparable price, a one order of magnitude higher peak performance than traditional CPUs. This drop in the cost of computation, as any…
Information seeking is a fundamental requirement for humans. However, existing LLM agents rely heavily on open-web search, which exposes two fundamental weaknesses: online content is noisy and unreliable, and many real-world tasks require…
Die-stacked DRAM is a promising solution for satisfying the ever-increasing memory bandwidth requirements of multi-core processors. Manufacturing technology has enabled stacking several gigabytes of DRAM modules on the active die, thereby…
This paper presents a games-in-games approach to provide design guidelines for mosaic command and control that enables the secure and resilient multi-domain operations. Under the mosaic design, pieces or agents in the network are equipped…
Task offloading in three-layer fog computing environments presents a critical challenge due to user equipment (UE) mobility, which frequently triggers costly service migrations and degrades overall system performance. This paper addresses…
Conventional cache models are not suited for real-time parallel processing because tasks may flush each other's data out of the cache in an unpredictable manner. In this way the system is not compositional so the overall performance is…
Disaggregation and rack-scale systems have the potential of drastically decreasing TCO and increasing utilization of cloud datacenters, while maintaining performance. While the concept of organising resources in separate pools and…
We present a single-node, multi-GPU programmable graph processing library that allows programmers to easily extend single-GPU graph algorithms to achieve scalable performance on large graphs with billions of edges. Directly using the…
Modern applications demand high performance and cost efficient database management systems (DBMSs). Their workloads may be diverse, ranging from online transaction processing to analytics and decision support. The cloud infrastructure…
Heterogeneous MPSoCs comprise diverse processing units of varying compute capabilities. To date, the mapping strategies of neural networks (NNs) onto such systems are yet to exploit the full potential of processing parallelism, made…
In this report we discuss the concepts of geometric memory align- ment, geometric memory allocation and geometric memory mapping. We introduce block trees as an efficient data structure for representing geo- metrically aligned block…
Hardware memory disaggregation (HMD) is an emerging technology that enables access to remote memory, thereby creating expansive memory pools and reducing memory underutilization in datacenters. However, a significant challenge arises when…
Processing large-scale graph datasets is computationally intensive and time-consuming. Processor-centric CPU and GPU architectures, commonly used for graph applications, often face bottlenecks caused by extensive data movement between the…
The deployment of mixture-of-experts (MoE) large language models (LLMs) presents significant challenges due to their high memory demands. These challenges become even more pronounced in multi-tenant environments, where shared resources must…
Radiology reports contain rich clinical information that can be used to train imaging models without relying on costly manual annotation. However, existing approaches face critical limitations: rule-based methods struggle with linguistic…
This paper presents Block, a distributed scheduling framework designed to optimize load balancing and auto-provisioning across instances in large language model serving frameworks by leveraging contextual information from incoming requests.…
GPUs are uniquely suited to accelerate (SQL) analytics workloads thanks to their massive compute parallelism and High Bandwidth Memory (HBM) -- when datasets fit in the GPU HBM, performance is unparalleled. Unfortunately, GPU HBMs remain…
The Vega grammar has been broadly adopted by a growing ecosystem of browser-based visualization tools. However, the reference Vega renderer does not scale well to large datasets (e.g., millions of rows or hundreds of megabytes) because it…