Related papers: MARS: Memory Aware Reordered Source
To improve efficiency, nearly all parallel processing units (CPUs and GPUs) implement relaxed memory models in which memory operations may be re-ordered, i.e., executed out-of-order. Prior testing work in this area found that memory…
Over the past two decades, the storage capacity and access bandwidth of main memory have improved tremendously, by 128x and 20x, respectively. These improvements are mainly due to the continuous technology scaling of DRAM (dynamic…
Recommendation models are very large, requiring terabytes (TB) of memory during training. In pursuit of better quality, the model size and complexity grow over time, which requires additional training data to avoid overfitting. This model…
The continued growth of the computational capability of throughput processors has made throughput processors the platform of choice for a wide variety of high performance computing applications. Graphics Processing Units (GPUs) are a prime…
The initial location of data in DRAMs is determined and controlled by the 'address-mapping' and even modern memory controllers use a fixed and run-time-agnostic address mapping. On the other hand, the memory access pattern seen at the…
Dynamic Random Access Memory (DRAM) is the de-facto choice for main memory devices due to its cost-effectiveness. It offers a larger capacity and higher bandwidth compared to SRAM but is slower than the latter. With each passing generation,…
We present the first systematic investigation of graph reordering effects for graph-based Approximate Nearest Neighbor Search (ANNS) on a GPU. While graph-based ANNS has become the dominant paradigm for modern AI applications, recent…
Optimization pipelines targeting polyhedral programs try to maximize the compute throughput. Traditional approaches favor reuse and temporal locality; while the communicated volume can be low, failure to optimize spatial locality may cause…
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…
Metaverse has become a buzzword recently. Mobile augmented reality (MAR) is a promising approach to providing users with an immersive experience in the Metaverse. However, due to limitations of bandwidth, latency and computational…
Unstructured-mesh based numerical algorithms such as finite volume and finite element algorithms form an important class of applications for many scientific and engineering domains. The key difficulty in achieving higher performance from…
GPGPU architectures have become established as the dominant parallelization and performance platform achieving exceptional popularization and empowering domains such as regular algebra, machine learning, image detection and self-driving…
Radio maps reflect the spatial distribution of signal strength and are essential for applications like smart cities, IoT, and wireless network planning. However, reconstructing accurate radio maps from sparse measurements remains…
Memory tiering has received wide adoption in recent years as an effective solution to address the increasing memory demands of memory-intensive workloads. However, existing tiered memory systems often fail to meet service-level objectives…
The exponential growth of data-intensive machine learning workloads has exposed significant limitations in conventional GPU-accelerated systems, especially when processing datasets exceeding GPU DRAM capacity. We propose MQMS, an augmented…
High level programming languages and GPU accelerators are powerful enablers for a wide range of applications. Achieving scalable vertical (within a compute node), horizontal (across compute nodes), and temporal (over different generations…
Management is a complex task in today's heterogeneous and large scale networks like Cloud, IoT, vehicular and MPLS networks. Likewise, researchers and developers envision the use of artificial intelligence techniques to create cognitive and…
Tree-based data structures are ubiquitous across applications. Therefore, a multitude of different tree implementations exist. However, while these implementations are diverse, they share a tree structure as the underlying data structure.…
We study the problem of spectrum sharing between goal-oriented (GO) and legacy data-oriented (DO) systems. For the former, data quality and representation is no longer optimized based on classical communication key performance indicators,…
Today's systems are overwhelmingly designed to move data to computation. This design choice goes directly against at least three key trends in systems that cause performance, scalability and energy bottlenecks: (1) data access from memory…