Scientific experiments are producing unprecedented volumes of data with real-time High Performance Computing (HPC) needs. Understanding and ensuring efficient data movement in these emerging data-intensive workloads is becoming critical for successful ...

Repartitioning in a parallel setting can be defined as the task of redistributing data across processes based on a newly imposed grid/layout. Repartitioning is a fundamental problem, with applications in domains that typically involve computation on ...

Hardware accelerators are used to speed up computationally expensive applications in many scientific fields. However, offloading tasks to accelerator cards requires data to be transferred between the memory of the host and the external memory of the ...

The new emerging scientific workloads to be executed in the upcoming exascale supercomputers face major challenges in terms of storage, given their extreme volume of data. In particular, intelligent data placement, instrumentation, and workflow handling ...

Traditional data centers include monolithic servers that tightly integrate CPU, memory and disk (Figure 1a). Instead, Disaggregated Systems (DSs) [8, 13, 18, 27] organize multiple compute (CC), memory (MC) and storage devices as independent, failure-...

Resource disaggregation offers a cost effective solution to resource scaling, utilization, and failure-handling in data centers by physically separating hardware devices in a server. Servers are architected as pools of processor, memory, and storage ...

With generational gains from transistor scaling, GPUs have been able to accelerate traditional computation-intensive workloads. But with the obsolescence of Moore's Law, single GPU systems are no longer able to satisfy the computational and memory ...

Due to the irregular memory access and high bandwidth demanding, graph processing is usually inefficient on conventional computer architectures. The recent development of the processing-in-memory (PIM) technique such as hybrid memory cube (HMC) has ...

Data movement is becoming the dominant contributor to the time and energy costs of computation across a wide range of application domains. However, time complexity is inadequate to analyze data movement. This work expands upon Data Movement Distance, a ...

Key-value databases are widely used to store the features or metadata generated from the neural network based video processing platforms. Due to the large volumes of video data, these databases use solid state drives (SSDs) as the primary data storage ...

Current systems hide data movement from software behind the load-store interface. Software's inability to observe and respond to data movement is the root cause of many inefficiencies, including the growing fraction of execution time and energy devoted ...

Conventional planar video streaming is the most popular application in mobile systems. The rapid growth of 360° video content and virtual reality (VR) devices is accelerating the adoption of VR video streaming. Unfortunately, video streaming consumes ...

Despite continuing research into inter-GPU communication mechanisms, extracting performance from multi-GPU systems remains a significant challenge. Inter-GPU communication via bulk DMA-based transfers exposes data transfer latency on the GPU's critical ...

Emerging multi-chip module GPUs (MCM-GPUs) expend over 17% of the total power budget on chip interconnects and this fraction is expected to increase as chip size increases. Towards proactively managing the power consumption of these interconnects, we ...

High GPU performance can only be achieved if a kernel efficiently uses the multi-layered compute and memory hierarchies. For example, accelerators such as NVIDIA ?s Tensor Cores require specific mappings of threads to data that must be considered in ...

With the end of Dennard scaling, power constraints have led to increasing compute specialization in the form of differently specialized accelerators integrated at various levels of the general-purpose system hierarchy. The result is that the most common ...

High-performance applications running on current and future architectures are mostly performance-limited by the cost of data movement, vertically through the memory hierarchy of a node or between CPU host and accelerator, and horizontally across nodes. ...

We propose ZnG, a new GPU-SSD integrated architecture, which can maximize the memory capacity in a GPU and address performance penalties imposed by an SSD. Specifically, ZnG replaces all GPU internal DRAMs with an ultra-low-latency SSD to maximize the ...

The cost of moving data between compute elements and storage elements plays a signiicant role in shaping the overall performance of applications.We present a compiler-driven approach to reducing data movement costs. Our approach, referred to as ...

Processing-In-Memory (PIM) architectures based on recent technology advances (e.g., Hybrid Memory Cube) demonstrate great potential for graph processing. However, existing solutions did not address the key challenge of graph processing---irregular data ...