Proactive task migration with a self-adjusting migration threshold for dynamic thermal management of multi-core processors

Salami, Bagher; Baharani, Mohammadreza; Noori, Hamid

doi:10.1007/s11227-014-1140-y

Cited by 27 publications

(8 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They used scalar temperature sensor measurements alone to derive two metrics that help decide whether to place workload on a server or not: the first metric, Thermal Correlation Index (TCI), gives the efficiency with which any given CRAC can provide cooling resources to any given server; and the second is the Local Workload Placement Index (LWPI). For multi-core processors, Baghaer et al [25] proposed selfadjusting temperature threshold schema for dynamic thermal management to minimize both average and peak temperature with low performance overhead. Tang et al [26] investigated the mechanism to distribute in-coming tasks among the servers in order to maximize cooling efficiency while still operating within safe temperature regions.…”

Section: State Of the Artmentioning

confidence: 99%

Model-Based Thermal Anomaly Detection in Cloud Datacenters Using Thermal Imaging

Lee

Viswanathan

Pompili

2018

IEEE Trans. Cloud Comput.

View full text Add to dashboard Cite

The growing importance, large scale, and high server density of high-performance computing datacenters make them prone to attacks, misconfigurations, and failures (of the cooling as well as of the computing infrastructure). Such unexpected events often lead to thermal anomalies -hotspots, fugues, and coldspots -which impact the cost of operation of datacenters. A modelbased thermal anomaly detection mechanism, which compares expected (obtained using heat-generation and -extraction models) and observed thermal maps (obtained using thermal cameras) of datacenters, is proposed. In addition, a novel Thermal Anomaly-aware Resource Allocation (TARA) is designed to induce a time-varying thermal fingerprint (thermal map) of the datacenter so to maximize the detection accuracy of the anomalies. As shown via experiments on a small-scale testbed as well as via trace-driven simulations, such model-based thermal anomaly detection solution in conjunction with TARA significantly improves the detection probability compared to anomaly detection when scheduling algorithms such as random, round robin, and best-fit-decreasing are employed.

show abstract

Section: State Of the Artmentioning

confidence: 99%

Model-Based Thermal Anomaly Detection in Cloud Datacenters Using Thermal Imaging

Lee

Viswanathan

Pompili

2018

IEEE Trans. Cloud Comput.

View full text Add to dashboard Cite

show abstract

“…They used scalar temperature sensor measurements alone to derive two metrics that help decide whether to place workload on a server or not: the first metric, Thermal Correlation Index (TCI), gives the efficiency with which any given CRAC can provide cooling resources to any given server; while the second is Local Workload Placement Index (LWPI). For multi-core processors, Baghaer et al [17] proposed self-adjusting temperature threshold schema for dynamic thermal management to minimize both average and peak temperature with low performance overhead.…”

Section: Management Of Heat Extractionmentioning

confidence: 99%

Proactive Thermal-Aware Resource Management in Virtualized HPC Cloud Datacenters

Lee

Viswanathan

Pompili

2017

IEEE Trans. Cloud Comput.

View full text Add to dashboard Cite

Clouds provide the abstraction of nearly-unlimited computing resources through the elastic use of federated resource pools (virtualized datacenters). They are being increasingly considered for HPC applications, which have traditionally targeted grids and supercomputing clusters. However, maximizing energy efficiency and utilization of cloud datacenter resources, avoiding undesired thermal hotspots (due to overheating of over-utilized computing equipment), and ensuring quality of service guarantees for HPC applications are all conflicting objectives, which require joint consideration of multiple pairwise tradeoffs. An innovative proactive thermal-aware virtual machine consolidation (involving allocations as well as migrations) technique is proposed to maximize computing resource utilization, to minimize datacenter energy consumption for computing, and to improve the efficiency of heat extraction. The capability to migrate virtual machines away from lightly-loaded servers in a thermal-aware manner opens up opportunity to improve resource consolidation over time and, hence, achieve the aforementioned goals. The effectiveness of the proposed technique is verified through experimental evaluations with HPC workload traces under single-as well as federated-datacenter scenarios. The trend: Datacenters are a growing component of society's IT infrastructure and their energy consumption surpassed 237 billion kWh/year worldwide and 76 billion kWh/year in the US in 2010 [2]. Even though these numbers are lower than what the US Environmental Protection Agency predicted in 2007 [3], they correspond to 6% and 2% of the total electricity usage in the US. The impact of this proliferation of datacenters on the environment and society includes increase in CO 2 emissions, overload of the electricity supply grid, and rise in water usage for cooling leading to water scarcity [4]. The scale and complexity of datacenters are growing at an alarming rate and their management is rapidly exceeding human ability, making autonomic (self-configuration, self-optimization, self-healing, and self-protection) management approaches essential.High-Performance Computing (HPC) applications are resource-intensive scientific workflow (in terms of data, computation, and communication) that have typically targeted Grids and conventional HPC platforms like super-computing clusters. Clouds -composed of one or more virtualized datacenters providing the abstraction of nearly-unlimited computing resources through the elastic use of federated resource pools -are being increasingly considered to enable traditional HPC applications. However, maximizing energy efficiency and utilization of cloud datacenter resources, avoiding undesired ther-

show abstract

“…Much previous work focused on phase-based tuning [8,15,16] and DTM [1][2][3]17]. Since we leverage both phase-based tuning and DTM, we present related work and background in these two areas.…”

Section: Background and Related Workmentioning

confidence: 99%

“…For each generation, TaPT uses the previous generation's Pareto optimal set as the current generation's initial archive (line 15). TaPT calculates the fitness of the configurations in each population and archive using Equations (2) and (3), and updates the current generation's archive with the non-dominated configurations (lines [17][18][19][20][21]. To maintain the size of P i 's archive at A size , TaPT discards the least fit configurations or adds the most fit configurations from the population (line 22).…”

Section: Tapt Characterization Algorithmmentioning

confidence: 99%

“…Input: n, s, A size , G, Q Output: P i 's best configuration 1: t ← 0 2: for i ← 1 to s do 3: C i ← rand()/s + 1 4: end for 5: population is C 1 , C 2 , ..., C s 6: for j ← 1 to n do 7: 10: if n == 0&&t == 0 then 11: archive ← ∅ 12: else if k > 0&&t == 0 then 13: archive ← A msp 14: else 15: archive ← archive(t − 1) 16: end if 17: U ← population + archive 18: for (C i ∈ U) do 19: fit(C i ) ← calculateFitness(C i ) 20: end for 21: archive ← getNonDominated(U) 22: size(archive) ← A size 23: if t == (G − 1) then 24: bestConfiguration(P i ) ← min( f (Q)) 25: else 26: t + + 27: go to line 2 28: end if When the final generation is reached, TaPT selects the best configuration from the archive that optimizes the designer-specified priority setting (line 24). Finally, TaPT stores C Pi in the phase history table ( Figure 4) for P i 's subsequent executions.…”

Section: Algorithm 1 Tapt Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

TaPT: Temperature-Aware Dynamic Cache Optimization for Embedded Systems

Adegbija

Gordon‐Ross

2017

Computers

View full text Add to dashboard Cite

Embedded systems have stringent design constraints, which has necessitated much prior research focus on optimizing energy consumption and/or performance. Since embedded systems typically have fewer cooling options, rising temperature, and thus temperature optimization, is an emergent concern. Most embedded systems only dissipate heat by passive convection, due to the absence of dedicated thermal management hardware mechanisms. The embedded system's temperature not only affects the system's reliability, but can also affect the performance, power, and cost. Thus, embedded systems require efficient thermal management techniques. However, thermal management can conflict with other optimization objectives, such as execution time and energy consumption. In this paper, we focus on managing the temperature using a synergy of cache optimization and dynamic frequency scaling, while also optimizing the execution time and energy consumption. This paper provides new insights on the impact of cache parameters on efficient temperature-aware cache tuning heuristics. In addition, we present temperature-aware phase-based tuning, TaPT, which determines Pareto optimal clock frequency and cache configurations for fine-grained execution time, energy, and temperature tradeoffs. TaPT enables autonomous system optimization and also allows designers to specify temperature constraints and optimization priorities. Experiments show that TaPT can effectively reduce execution time, energy, and temperature, while imposing minimal hardware overhead.

show abstract

Proactive task migration with a self-adjusting migration threshold for dynamic thermal management of multi-core processors

Cited by 27 publications

References 16 publications

Model-Based Thermal Anomaly Detection in Cloud Datacenters Using Thermal Imaging

Model-Based Thermal Anomaly Detection in Cloud Datacenters Using Thermal Imaging

Proactive Thermal-Aware Resource Management in Virtualized HPC Cloud Datacenters

TaPT: Temperature-Aware Dynamic Cache Optimization for Embedded Systems

Contact Info

Product

Resources

About