Browsing by Subject "Energy Efficiency"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Architectural Exploration of Data Recomputation for Improving Energy Efficiency
    (2017-07) Akturk, Ismail
    There are two fundamental challenges for modern computer system design. The first one is accommodating the increasing demand for performance in a tight power budget. The second one is ensuring correct progress despite the increasing possibility of faults that may occur in the system. To address the first challenge, it is essential to track where the power goes. The energy consumption of data orchestration (i.e., storage, movement, communication) dominates the energy consumption of actual data production, i.e., computation. Oftentimes, recomputing data becomes more energy efficient than storing and retrieving pre-computed data by minimizing the prevalent power and performance overhead of data storage, retrieval, and communication. At the same time, recomputation can reduce the demand for communication bandwidth and shrink the memory footprint. In the first half of the dissertation, the potential of data recomputation in improving energy efficiency is quantified and a practical recomputation framework is introduced to trade computation for communication. To address the second challenge, it is needed to provide scalable checkpointing and recovery mechanisms. The traditional method to recover from a fault is to periodically checkpoint the state of the machine. Periodic checkpointing of the machine state makes rollback and restart of execution from a safe state possible upon detection of a fault. The energy overhead of checkpointing, however, as incurred by storage and communication of the machine state grows with the frequency of checkpointing. Amortizing this overhead becomes especially challenging, considering the growth of expected error rates as an artifact of contemporary technology scaling. Recomputation of data (which otherwise would be read from a checkpoint) can reduce both the frequency of checkpointing, the size of the checkpoints and thereby mitigate checkpointing overhead. In the second half, quantitative characterization of recomputation-enabled checkpointing (based on recomputation framework) is provided.
  • Thumbnail Image
    Item
    Impact of External Mechanisms on Energy Efficiency Investments at Small-and-medium-sized Suppliers
    (2018-02) Nguyen, Jason Quang
    The manufacturing sector is the world’s largest energy consumer, responsible for more than one third of the total final energy consumption of the global economy, and is com- prised of many small and medium-sized manufacturers (SMM). Governments, large cor- porations and third-party agencies have dedicated tremendous resources to regulatory policies and incentive mechanisms to encourage SMMs to invest in Energy Efficiency (EE) improvements. However, significant EE improvement opportunities at SMMs re- main unimplemented. In order to fully capitalize these potentials, it is thus essential to develop a better understanding of the investment decisions faced by SMMs. Previ- ous research has mainly considered EE investment decisions by an isolated single firm. However, for SMMs, it is also important to account for the potential influence that sup- ply chain interactions with large, powerful buyers and competing suppliers may have on such investment decisions. In the three studies of this research, we examine the impact of mechanisms governed by regulators, buyers and third-party agencies on the EE investment decisions of SMMs, taking into consideration supply chain interactions with both their buyers and competing suppliers. In the first study, we investigate when it is beneficial for a buyer to offer EE instru- ments, including assessment assistance and procurement commitment, and how these interact with third-party assessment assistance to impact the supplier’s EE investment level. We find that assessment assistance helps reduce the EE gap but procurement commitment is required to eliminate it. However, the buyer offers procurement com- mitment only when the alternate supplier is sufficiently expensive. Not surprisingly, third-party assessment assistance is important for unlocking EE improvement when the assessment cost is high. Nevertheless, when the costs of both the assessment and the al- ternate supplier are moderate, the addition of third-party assistance can actually harm EE investment by deterring the buyer from offering her own instruments. Energy mar- ket characteristics influence these outcomes in several ways. We find that an increase in the volatility or cross period correlation of energy prices reduces the buyer’s incentive to offer both assessment assistance and procurement commitment, leading to lower EE investment. However, an increase in the expected energy price generally increases the buyer’s incentive to offer both instruments, expanding the regions where the EE gap is reduced or even eliminated. In the second study, we examine the impact of carbon pricing on social welfare, taking into consideration the negative externality of energy and the positive externality generated by domestic manufacturing. We show that in the absence of external juris- diction competition, the first-best social welfare is achieved by setting the carbon price at the negative externality of energy. These results continue to hold in the presence of external jurisdiction competition, but only when the external costs of energy are suffi- ciently low. Setting a carbon price at the negative externality of energy when it is high is no longer optimal and the first-best outcome social welfare is not achieved. When so- cial welfare losses happen, neither of the two common remedies, carbon price relief and EE investment subsidy, can singlehandedly restore the losses. We show that a balanced combination of both remedies is required to achieve the first-best social welfare level. In the third study, we shift the focus to settings where a buyer is not aware of the supplier’s EE improvement opportunities. For example, western buyers like Wal-mart or Target are typically not well informed about the EE improvement potentials of their Chinese suppliers and need engagement from third-party agencies to learn the informa- tion. We analyze the conditions under which a third-party agency should work only with a supplier or also engage the buyer to achieve the higher EE investment from the supplier, as well as the impact of the third-party agency’s tactic on the supplier’s prof- itability. We find that buyer engagement can either help or hinder the supplier’s EE investment level, depending on the cost of the buyer’s alternative supply source. We also find that the potential benefit of engaging the buyer on the supplier’s EE invest- ment is reduced as the alternate supply source becomes more expensive. Regarding the impact on the supplier’s profitability, we show that buyer engagement always reduces the supplier’s profit as the engaged buyer squeezes all the associated EE cost savings. We further analyze the impact of energy market uncertainty on these directional re- sults. In particular, an increase in the volatility or cross period correlation of energy prices reduces the chance that third-party agency’s engagement with the buyer posi- tively influences the supplier’s EE investment level. However, when an improvement in EE investment does occur, its magnitude is larger for higher volatility or cross period correlation of energy prices. A higher volatility or cross period correlation of energy prices also reduces the detrimental impact of buyer engagement on the supplier’s profit. Our findings provide insights for policy makers interested in increasing EE investment and reducing the energy efficiency gap that plagues many supply chains.
  • Thumbnail Image
    Item
    Looking Beyond Demand Response: Barriers and Opportunities to Deploying Virtual Power Plants among Rural Electric Cooperatives in the United States
    (2024-05-16) Datta, Mayukh K.
    Rural electric cooperatives (co-ops) find themselves in a unique position regarding deploying virtual power plants. Co-ops, which are consumer-owned utilities, have a vast history of deploying controllable demand-side management technologies that can fit perfectly into a VPP framework, with almost a gigawatt of demand-side management capacity across four generation and transmission cooperatives in Minnesota (G. Chan et al. 2019; Matthew Grimley and Chan 2023). This more than forty-year-long experience deploying controllable resources and their nonprofit, consumer-owned structure makes rural electric cooperatives perfectly positioned to deploy virtual power plants. However, several challenges, such as high upfront costs and uncertainties around market rules, hinder VPP deployment for rural co-ops. Furthermore, the fact that most co-ops comprise a complex network of distribution cooperatives that make up larger generation and transmission (G&T) cooperatives also complicates how VPPs can be deployed by rural coops.
  • Thumbnail Image
    Item
    Power Consumption of Virtual Machines in Cloud Computing: Measurement and Enhancement
    (2016-07) BAI, YAN
    Virtualization is one of the cornerstone technologies that makes utility computing platforms such as cloud computing a reality. With the accelerating adoption of cloud computing, the virtualizaion-based cloud platforms are consuming a significant amount of energy. However, the design of a green and efficient virtualization technology remains an open issue to both industry and academia. In this thesis, we for the first time investigate the virtual machine's (VM's) power consumption while supporting different services and applications (e.g., web, database and streaming). In particular, we establish a cloud computing platform in the The University of Minnesota Duluth. This platform consist of both Xen and KVM nodes and the VMs can be easily accessed from the Internet. Our real-world measurement indicates that the existing virtulization technologies add considerable energy overhead to the data centers. For example, a busy virtualized database server can consume 30\% more energy than its non-virtualized counterparts. To address such a problem, we propose a shared-memory-based enhancement to reduces the extra interrupts and memory copies for cloud virtualization. The evaluation indicates that our approach can reduce VM's energy consumption by 11% without noticeable loss of its running performance.