Citation Context

...ng tries to address the effects of heat recirculation, a common phenomenon which is responsible for much of the energy inefficiency in contemporary data centers. Algorithms such as MinHR [2] and XInt =-=[3]-=- perform thermal-aware spatial job scheduling and manage to reduce the heat recirculation; they can effectively reduce the Supply Heat Index [14], a metric that describes the energy efficiency of a da...

Citation Context

...]. Most of the existing approaches for job scheduling in data centers focus exclusively on the job distribution between computing servers [7] targeting energy-efficient [8] or thermalaware scheduling =-=[9]-=-. To the best of our knowledge, only a few approaches have considered data center network and traffic characteristics for developing energy-efficient data center schedulers [10-12]. Ref. [10] identifi...

Citation Context

...derable effect on the energy efficiency of a data center. Thermal awareness, though, takes into account these phenomena, and there have been studies that enhance HPC scheduling with thermal awareness =-=[2, 3, 11]-=-. Typically, HPC and Internet data centers have similar physical layouts, therefore presumably they exhibit the same thermal phenomena. The approach in this paper aims to improve the aforementioned ac...

Citation Context

...ntial for energy saving as compared to Mol3D because of the lower MFLOP/s. 8. RELATED WORK Most researchers from HPC have focused on minimizing machine energy consumption as opposed to cooling energy =-=[15, 1, 21]-=-. Given a target program, a DVFS enabled cluster, and constraints on power consumption, they [18] come up with a frequency schedule that minimizes execution time while staying within the power constra...

Citation Context

... from high performance computing communities. The most elaborate thermal aware schedule algorithms for tasks in data centers are with computational fluid dynamics (CFD) models [5], [6]. Some research =-=[7]-=-, [8] declares that the CFD based model is too complex and is not suitable for online scheduling. This has lead to the development of some less complex online scheduling algorithms. Sensor-based fast ...

Citation Context

...maximum battery life. J. Wei et al and C. J. Xue et al [17, 18] develop an optimal lazy scheduler to manage services with minimum energy expenditure while not violating time-sensitive constraints. In =-=[19]-=-, a peak inlet temperature minimization problem is formulated to improve the energy efficiency. J. R. Cao et al [20] present a clustering architecture in order to obtain good performance in energy eff...

Citation Context

...aches for data center thermal management. Thermodynamics-based solutions derive thermal models of different locations inside a data center using fundamental thermodynamics laws and data center layout =-=[1, 21, 26, 31]-=-. On the other hand, data-centric solutions use data-mining [27] or machine learning algorithms [22] to model and optimize cooling in a data center. All these existing solutions, however, provide stat...

Citation Context

...the blade systems and performing linear regressions on the data, the idle chassis power consumption (ω) and the single fully-utilized server power consumption (α) values for the simulation runs [23], =-=[24]-=-, as given in Table II were computed. The simulations assume that the jobs are CPU-intensive. The estimated power consumption of the resulting linear function has an error of 0.4–9% from the actual me...

Citation Context

...lectricity bill of large datacenters [8]. There have been a number of prior works on increasing the efficiency of the cooling process in datacenters by performing temperature-aware task placement [9] =-=[10]-=-. In [10] the authors formulate and solve a mathematical problem that maximizes the steady state datacenter cooling efficiency by maximizing the required supplied cold-air temperature value. We used a...