Related papers: Taiji: A DPU Memory Elasticity Solution for In-pro…
In recent years, the use of artificial intelligence on resource-constrained IoT devices has grown significantly. However, existing approaches to DNN partitioning and offloading across the edge-cloud continuum typically rely on static…
Memory management is necessary with the increasing number of multi-connected AI devices and data bandwidth issues. For this purpose, high-speed multi-port memory is used. The traditional multi-port memory solutions are hard-bounded to a…
The explosive growth of artificial intelligence (AI) workloads in modern data centers demands a radical transformation of interconnect architectures. Traditional copper and optical wiring face fundamental challenges in latency, power…
Virtualization technology facilitates a dynamic, demand-driven allocation and migration of servers. This paper studies how the flexibility offered by network virtualization can be used to improve Quality-of-Service parameters such as…
Cloud Computing researches involve a tremendous amount of entities such as users, applications, and virtual machines. Due to the limited access and often variable availability of such resources, researchers have their prototypes tested…
Edge computing has emerged as a distributed computing paradigm to overcome practical scalability limits of cloud computing. The main principle of edge computing is to leverage on computational resources outside of the cloud for performing…
Shuffle exchanges intermediate results between upstream and downstream operators in distributed data processing and is usually the bottleneck due to factors such as small random I/Os and network contention. Several systems have been…
In this thesis, we describe a new, practical approach to integrating hardware-based data compression within the memory hierarchy, including on-chip caches, main memory, and both on-chip and off-chip interconnects. This new approach is fast,…
Diffusion Transformers (DiT) have established a new state-of-the-art in high-fidelity image synthesis; however, their massive computational complexity and memory requirements hinder local deployment on resource-constrained edge devices. In…
Serverless computing promises enhanced resource efficiency and lower user costs, yet is burdened by a heavyweight, CPU-bound data plane. Prior efforts exploiting shared memory reduce overhead locally but fall short when scaling across…
Today's Cloud applications are dominated by composite applications comprising multiple computing and data components with strong communication correlations among them. Although Cloud providers are deploying large number of computing and…
Deep learning (DL) jobs use multi-dimensional parallelism, i.e. combining data, model, and pipeline parallelism, to use large GPU clusters efficiently. Long-running jobs may experience changes to their GPU allocation: (i) resource…
With deep neural networks (DNNs) emerging as the backbone in a multitude of computer vision tasks, their adoption in real-world applications broadens continuously. Given the abundance and omnipresence of smart devices in the consumer…
Emerging hybrid memory systems that comprise technologies such as Intel's Optane DC Persistent Memory, exhibit disparities in the access speeds and capacity ratios of their heterogeneous memory components. This breaks many assumptions and…
Virtualization technology has enabled applications to be decoupled from the underlying hardware providing the benefits of portability, better control over execution environment and isolation. It has been widely adopted in scientific grids…
Adapting the cloud for high-performance computing (HPC) is a challenging task, as software for HPC applications hinges on fast network connections and is sensitive to hardware failures. Using cloud infrastructure to recreate conventional…
Virtualization is a promising technology that has facilitated cloud computing to become the next wave of the Internet revolution. Adopted by data centers, millions of applications that are powered by various virtual machines improve the…
Despite the high computational throughput of GPUs, limited memory capacity and bandwidth-limited CPU-GPU communication via PCIe links remain significant bottlenecks for accelerating large-scale data analytics workloads. This paper…
It is often said that one of the biggest limitations on computer performance is memory bandwidth (i.e."the memory wall problem"). In this position paper, I argue that if historical trends in computing evolution (where growth in available…
Serverless Large Language Models (LLMs) have emerged as a cost-effective solution for deploying AI services by enabling a 'pay-as-you-go' pricing model through GPU resource sharing. However, cold-start latency, especially the model loading…