Related papers: S/C: Speeding up Data Materialization with Bounded…

Deploying Small Language Models (SLMs) on edge platforms is critical for real-time, privacy-sensitive generative AI, yet constrained by memory, latency, and energy budgets. Quantization reduces model size and cost but suffers from device…

Modern Mixed-Criticality Systems (MCSs) rely on hardware heterogeneity to satisfy ever-increasing computational demands. However, most of the heterogeneous co-processors are designed to achieve high throughput, with their…

Accurate simulation techniques are indispensable to efficiently propose new memory or architectural organizations. As implementing new hardware concepts in real systems is often not feasible, cycle-accurate simulators employed together with…

Modern big data frameworks (such as Hadoop and Spark) allow multiple users to do large-scale analysis simultaneously. Typically, users deploy Data-Intensive Workflows (DIWs) for their analytical tasks. These DIWs of different users share…

DIMM-compatible persistent memory unites memory and storage. Prior works utilize persistent memory either by combining the filesystem with direct access on memory mapped files or by managing it as a collection of objects while abolishing…

Virtual machine (VM) scheduling is an important technique to efficiently operate the computing resources in a data center. Previous work has mainly focused on consolidating VMs to improve resource utilization and thus to optimize energy…

Differential computation (DC) is a highly general incremental computation/view maintenance technique that can maintain the output of an arbitrary and possibly recursive dataflow computation upon changes to its base inputs. As such, it is a…

Structured sparsity has been proposed as an efficient way to prune the complexity of Machine Learning (ML) applications and to simplify the handling of sparse data in hardware. Accelerating ML models, whether for training, or inference,…

We propose WSMC, a workload-specific memory capacity configuration approach for the Spark workloads, which guides users on the memory capacity configuration with the accurate prediction of the workload's memory requirement under various…

Approximate memory is a technique to mitigate the performance gap between memory subsystems and CPUs with its reduced access latency at a cost of data integrity. To gain benefit from approximate memory for realistic applications, it is…

Containerization plays a crucial role in the de facto technology stack for implementing microservices architecture (each microservice has its own database in most cases). Nevertheless, there are still fierce debates on containerizing…

Modern machine learning training is increasingly bottlenecked by data I/O rather than compute. GPUs often sit idle at below 50% utilization waiting for data. This paper presents a machine learning approach to predict I/O performance and…

In this research project, we investigate an alternative to the standard cloud-centralized data architecture. Specifically, we aim to leave part of the application data under the control of the individual data owners in decentralized…

In the era of burgeoning data generation, managing and storing large-scale time-varying datasets poses significant challenges. With the rise of supercomputing capabilities, the volume of data produced has soared, intensifying storage and…

In database query processing, actual run-time conditions (e.g., actual selectivities and actual available memory) very often differ from compile-time expectations of run-time conditions (e.g., estimated predicate selectivities and…

Hyperdimensional Computing (HDC) encodes information and data into high-dimensional distributed vectors that can be manipulated using simple bitwise operations and similarity searches, offering parallelism, low-precision hardware…

With recent video object segmentation (VOS) benchmarks evolving to challenging scenarios, we revisit a simple but overlooked strategy: restricting the size of memory banks. This diverges from the prevalent practice of expanding memory banks…

Magnetic random-access memory (MRAM) is a promising memory technology due to its high density, non-volatility, and high endurance. However, achieving high memory fidelity incurs significant write-energy costs, which should be reduced for…

We present SCM (Sleep-Consolidated Memory), a research preview of a memory architecture for large language models that draws on neuroscientific principles to address a fundamental limitation in current systems: the absence of persistent,…

Compute in-memory (CIM) is a promising technique that minimizes data transport, the primary performance bottleneck and energy cost of most data intensive applications. This has found wide-spread adoption in accelerating neural networks for…