On the impact of energy-saving strategies in opportunistic grids

Abstract

Opportunistic grids are distributed computing infrastructures that harvest the idle computing cycles of computing resources geographically distributed. In these grids, the demand for resources is typically bursty. During bursts of resource demand, many grid resources are required, but on other times they remain idle for long periods. If the resources are kept powered on even when they are neither processing their owners workload nor grid jobs, their exploitation is not efficient in terms of energy consumption. One way to reduce the energy consumed in these idleness periods is to place the computers that form the grid in a "sleeping" mode which consumes less energy. We evaluated two sleeping strategies, denoted: standby and hibernate. Resources that comprise an opportunistic grid are normally very heterogeneous, and differ enormously on their processing power and energy consumption. It opens the possibility of implementing scheduling strategies that take energy-efficiency into account. We consider scheduling in two different levels. Firstly, how to choose which machine should be woken up, if several options are available. Secondly, how to decide which tasks to schedule to the available machines. In summary, our results presented a significant reduction in energy consumption, surpassing 80% in a scenario when the amount of resources in the grid was high. Moreover, this comes with limited impact on the response time of the applications.

Modeling and simulation of global and sleep states in ACPI‐compliant energy‐efficient cloud environments

“…This new experiment expands our validations to a scenario that covers other states presented in this work. For this purpose, we evaluated a well‐known timeout‐based strategy that uses four power states ( V ={ G 0, S 3, S 4, G 2}): When the host in the G0 state becomes idle, it enters into the S3 state; The host returns immediately to the G0 state if it is requested; and The host enters successively to a lower‐power state if the timeout expires. …”

“…This new experiment expands our validations to a scenario that covers other states presented in this work. For this purpose, we evaluated a well-known timeout-based strategy [33][34][35] that uses four power states (V D ¹G0; S3; S4; G2º):…”

Trace-Driven Simulation for Energy Consumption in High Throughput Computing Systems

“…Minartz et al [12] explore energy saving through the dynamic provisioning of resources within a high performance computing cluster, while Niemi et al [13] demonstrate energy savings through the consolidation of multiple jobs onto the same hardware. Ponciano et al evaluate strategies for energy-aware resource provisioning and job allocation within opportunistic grids [14]. Terzopoulos et al investigate the use of Dynamic Voltage Scaling techniques to reduce energy consumption in a heterogeneous cluster to conform to power budgets [15] imposed by infrastructure.…”

HTC‐Sim: a trace‐driven simulation framework for energy consumption in high‐throughput computing systems

“…Ponciano et al [30] evaluate strategies for energy-aware resource provisioning and job allocation within opportunistic grids, transitioning worker nodes into energy-saving sleep modes during idle periods. Zikos et al [31] model a cluster within a computational grid as an open queueing network and evaluate the impact of resource allocation strategies on performance and energy consumption.…”