Search Results

In this paper we investigate performance-energy optimization of tokenizer algorithm training using power capping. We focus on parallel, multi-threaded implementations of Byte Pair Encoding (BPE), Unigram, WordPiece, and WordLevel run on two ...

In this paper we demonstrate that it is possible to obtain considerable improvement of performance and energy aware metrics for training of deep neural networks using a modern parallel multi-GPU system, by enforcing selected, non-default power ...

Given the increasing complexity and the heterogeneity of today's computing system infrastructure, power efficiency and fault tolerance remain the top challenges of an High Performance Computing (HPC) facility operation. Recently, many research efforts ...

Cloud providers, like Amazon and Microsoft, must guarantee high availability for a large fraction of their workloads. For this reason, they build datacenters with redundant infrastructures for power delivery and cooling. Typically, the redundant ...

While transitioning to exascale systems, it has become clear that power management plays a fundamental role to support a viable utilization of the underlying hardware, also performance-wise. To meet power restrictions imposed by future exascale ...

As major web and cloud service providers continue to accelerate the demand for new data center capacity worldwide, the importance of power oversubscription as a lever to reduce provisioning costs has never been greater. Building on insights from Google-...

As the scale and density of data centers continue to grow, cost-effective data center management (DCM) is becoming a significant challenge for enterprises hosting large-scale online and cloud services. Machines need to be monitored, and the scale of ...

The ongoing evolution of the power grid towards a highly dynamic supply system poses challenges as renewables induce new grid characteristics. The volatility of electricity sources leads to a fluctuating electricity price, which even becomes negative ...

Recent trends in data centers include on one hand consolidation in large data centers and on the other hand increased number of small edge systems located closer to sources of data or energy. In both cases the key features needed to minimise costs are ...

Power and energy consumption are now key design concerns in HPC. To develop software that meets power and energy constraints, scientific application developers must have a reliable way to measure these values and relate them to application-specific ...

A future large-scale high-performance computing (HPC) cluster will likely be power capped since the surrounding infrastructure like power supply and cooling is constrained. For such a cluster, it may be impossible to supply thermal design power (TDP) to ...

In this manuscript we evaluate the impact of HW power capping mechanisms on a real scientific application composed by parallel execution. By comparing HW capping mechanism against static frequency allocation schemes we show that a speed up can be ...

In current computing systems, many applications require guarantees on their maximum power consumption to not exceed the available power budget. On the other hand, for some applications, it could be possible to decrease their performance, yet maintain an ...

Power consumption of the world's leading supercomputers is of the order of tens of MegaWatts (MW). Therefore, energy efficiency and power management of High Performance Computing (HPC) systems are among the main goals of the HPC community. This paper ...

Power consumption of extreme-scale supercomputers has become a key performance bottleneck. Yet current practices do not leverage power management opportunities, instead running at "maximum power". This is not sustainable. Future systems will need to ...

Power shaving has recently been proposed to dynamically shave the power peaks of a data center with energy storage devices (ESD), such that more servers can be safely hosted. In addition to the reduction of capital investment (cap-ex), power shaving also ...

Constraints imposed by power delivery and costs will be key design impediments to the development of next generation High-Performance Computing (HPC) systems. To remedy these impediments, solutions that impose power bounds (or caps) on over-provisioned ...

Online defect screening techniques to detect run-time faults are becoming a necessity in current and near future technologies. At the same time, due to aggressive technology scaling into the nanometer regime, power consumption is becoming a significant ...

This paper presents a power-aware scheduling algorithm based on efficient distribution of the computing workload to the resources on heterogeneous CPU-GPU architectures. The scheduler manages the resources of several computing nodes with a view to ...

This paper studies a multiobjective evolutionary game theoretic framework for application placement in clouds that support a power capping mechanism (e.g., Intel's Runtime Average Power Limit--RAPL) for CPUs. Given the notion of power capping, power can ...