Related papers: Pond: CXL-Based Memory Pooling Systems for Cloud P…

Efficient custom pooling techniques that can aggressively trim the dimensions of a feature map and thereby reduce inference compute and memory footprint for resource-constrained computer vision applications have recently gained significant…

Neuroimaging open-data initiatives have led to increased availability of large scientific datasets. While these datasets are shifting the processing bottleneck from compute-intensive to data-intensive, current standardized analysis tools…

Cloud providers offer a variety of execution platforms in form of bare-metal, VM, and containers. However, due to the pros and cons of each execution platform, choosing the appropriate platform for a specific cloud-based application has…

Heterogeneous memory technologies are increasingly important instruments in addressing the memory wall in HPC systems. While most are deployed in single node setups, CXL.mem is a technology that implements memories that can be attached to…

Cloud providers have introduced pricing models to incentivize long-term commitments of compute capacity. These long-term commitments allow the cloud providers to get guaranteed revenue for their investments in data centers and computing…

As users migrate their analytical workloads to cloud databases, it is becoming just as important to reduce monetary costs as it is to optimize query runtime. In the cloud, a query is billed based on either its compute time or the amount of…

New dynamic cloud pricing options are emerging with cloud providers offering resources as a wide range of CPU frequencies and matching prices that can be switched at runtime. On the other hand, cloud providers are facing the problem of…

Cloud resource management is often modeled by two-dimensional bin packing with a set of items that correspond to tasks having fixed CPU and memory requirements. However, applications running in clouds are much more flexible: modern…

The process of pooling vertices involves the creation of a new vertex, which becomes adjacent to all the vertices that were originally adjacent to the endpoints of the vertices being pooled. After this, the endpoints of these vertices and…

Message Passing Interface (MPI) is a foundational programming model for high-performance computing. MPI libraries traditionally employ network interconnects (e.g., Ethernet and InfiniBand) and network protocols (e.g., TCP and RoCE) with…

Cloud computing is a particular implementation of distributed computing. It inherited many properties of distributed computing such as scalability, reliability and distribution transparency. The transparency middle layer abstracts the…

The growing demands in the training and inference of Large Language Models (LLMs) are accelerating the adoption of scale-up systems that extend server shared memory through the use of Compute Express Link (CXL)-based load/store…

Data movement in memory-intensive workloads, such as deep learning, incurs energy costs that are over three orders of magnitude higher than the cost of computation. Since these workloads involve frequent data transfers between memory and…

Nowadays, Deep Neural Networks are among the main tools used in various sciences. Convolutional Neural Network is a special type of DNN consisting of several convolution layers, each followed by an activation function and a pooling layer.…

Prefix caching is crucial to accelerate multi-turn interactions and requests with shared prefixes. At the cluster level, existing prefix caching systems are tightly coupled with request scheduling to optimize cache efficiency and…

The cloud computing paradigm underlines data center and telecommunication infrastructure design. Heavily leveraging virtualization, it slices hardware and software resources into smaller software units for greater flexibility of…

3D point cloud neural networks have significantly enhanced the perceptual capabilities of resource-limited mobile intelligent systems. However, despite the transformative impact, the point cloud algorithm suffers from substantial memory…

Processing-in-memory (PIM) reduces data movement by executing near memory, but our large-scale characterization on real PIM hardware shows that end-to-end performance is often limited by disjoint host and device address spaces that force…

Pooling is one of the main elements in convolutional neural networks. The pooling reduces the size of the feature map, enabling training and testing with a limited amount of computation. This paper proposes a new pooling method named…

Computing has a huge memory problem. The memory system, consisting of multiple technologies at different levels, is responsible for most of the energy consumption, performance bottlenecks, robustness problems, monetary cost, and hardware…