Energy Efficient Configuration for QoS in Reliable Parallel Servers

Abstract

Abstract. Redundancy is the traditional technique used to increase system reliability. With modern technology, in addition to being used as temporal redundancy, slack time can also be used by energy management schemes to scale down system processing speed and supply voltage to save energy. In this paper, we consider a system that consists of multiple servers for providing reliable service. Assuming that servers have self-detection mechanisms to detect faults, we first propose an efficient parallel recovery scheme that processes service requests in parallel to increase the number of faults that can be tolerated and thus the system reliability. Then, for a given request arrival rate, we explore the optimal number of active severs needed for minimizing system energy consumption while achieving k-fault tolerance or for maximizing the number of faults to be tolerated with limited energy budget. Analytical results are presented to show the trade-off between the energy savings and the number of faults being tolerated.

Collaboration among mobile agents for efficient energy allocation in mobile grid

“…The iterative algorithm that achieves collaboration algorithm among mobile agents for efficient energy allocation (CAMA) is described as follows. (Zhu et al 2005;Kim et al 2005). Assume that the maximum power, P max corresponds to running all jobs with the maximum processing frequency.…”

“…Energy efficiency for communication system and computer network has recently become hot research area (Zhou et al 2004;Huang and Kao 2007;Chen and Leneutre 2008;Alsalih et al 2005;Shang et al 2002). Energy and deadline constrained scheduling have been studied in embedded system (Kumar et al 2007), cluster system (Seshasayee et al 2007;Zong and Qin 2007;Huang et al 2005;Kim et al 2005Kim et al , 2007Zhu et al 2005) and real-time system (Kumar et al 2007), but considering energy in grid computing is few.…”

Design Optimization of Time- and Cost-Constrained Fault-Tolerant Embedded Systems With Checkpointing and Replication

“…Multiple failures are addressed with active replication in [13] in order to guarantee a required level of fault tolerance and satisfy time constraints. [43] use software replication and temporal redundancy to provide a tradeoff between energy savings and the number of faults being tolerated in the context of reliable parallel servers.…”

Reliability-Aware Dynamic Energy Management in Dependable Embedded Real-Time Systems

“…However, considering the various factors that affect the transient fault rate (such as cosmic ray flux, technology feature size, chip capacity, supply voltage and operating frequency), obtaining a 1 We note that this conclusion has been also reached by several research groups, though through different energy modeling techniques [8,11,17,13,28]. 2 For systems with multiple processing units, some processing units may be powered off for energy efficiency and the energy savings will be affected by P s [36,42]. 3 For discrete frequency levels, we can use two adjacent levels to emulate the execution at any frequency [14].…”

“…That is, we assume the working system is always on and focus on system active power. For the effects of different values of α and β on energy management, see [42,43] for more discussions. The same as in Section 3, we assume that faults follow a Poisson distribution with an average fault rate as λ 0 = 10 −6 at f max (and corresponding V max ).…”