Mountain Scholar

Mountain Scholar is an open access repository service that collects, preserves, and provides access to digitized library collections and other scholarly and creative works from Colorado State University and the University Press of Colorado. It also serves as a dark archive for the Open Textbook Library.

Communities in Mountain Scholar

Select a community to browse its collections.

Now showing 1 - 3 of 3

Colorado State University, Fort Collins 29067
Explore the Colorado State University community’s scholarly output as well as items from the University at large and the CSU Libraries.
Open Textbook Library Archive 1866
A limited number of titles are available here. To see all OTL titles, please visit the Open Textbook Library at https://open.umn.edu/opentextbooks. Only Open Textbook Library staff have access to all OTL Archive titles held in Mountain Scholar.
University Press of Colorado 1106
Access is limited to University Press of Colorado members. Non-members: to purchase books, please visit https://upcolorado.com/.

Recent Submissions

Access status: Open Access ,
Sustainable graph analytics workload scheduling with evolutionary reinforcement learning in edge-cloud systems
(2026-06-22) Ramicetty, Pallavi, author; Moore, Hayden, author; Qi, Sirui, author; Islam, Akhirul, author; Ghose, Manojit, author; Milojicic, Dejan, author; Bash, Cullen, author; Pasricha, Sudeep, author; ACM, publisher
Graph analytics powers modern intelligent systems such as smart cities, cyber-physical infrastructure, IoT security, and large-scale social networks. As these workloads scale in complexity, their execution in heterogeneous edge-cloud environments results in higher energy use and carbon emission footprint. To address this challenge, we propose MERSEM, a multi-objective evolutionary reinforcement learning framework for sustainable edge-cloud system management. MERSEM integrates evolutionary search with reinforcement learning (RL) to solve the problem of graph workload allocation and scheduling. The evolutionary component explores diverse global solutions, while the RL agent refines decisions through adaptive local optimization. The framework is designed to jointly minimize service-level agreement (SLA) violations and carbon emissions by considering dynamic carbon intensity, resource heterogeneity, and workload characteristics. Experimental results demonstrate that MERSEM outperforms the state-of-the-art with up to 45% SLA violation reductions and up to 12% carbon emission reductions.
Access status: Open Access ,
Revisiting "cooler is better": ITD-aware per-CPU thermal optimization for sustainable data center operation
(2026-06-22) Crop, Jason, author; Moore, Hayden, author; Pasricha, Sudeep, author; ACM, publisher
As data center energy demand approaches grid-level constraints, optimizing conventional server infrastructure is essential for sustainable growth. The long-standing assumption that "cooler is better", i.e., lower CPU temperatures reduce power, does not fully hold for modern low-voltage CPUs, where inverse temperature dependence (ITD) drives higher supply voltages at lower temperatures. This creates a non-monotonic performance-per-watt curve where efficiency peaks at an intermediate thermal point. In this paper, for the first time, we empirically characterize ITD on production Intel Xeon CPUs and demonstrate that efficiency-optimal temperatures are CPU part-specific, and frequently higher than typical data center operating conditions. Measurements from commercial cloud data center platforms (Amazon, Equinix) reveal that approximately half of modern high-power CPUs operate about 10°C below their efficiency-optimal thermal point. By implementing ITD-aware thermal grouping of CPUs and inlet temperature adjustments, data center operators can optimize facility-level cooling and overall sustainability. Our case study shows that this approach can reduce total data center energy by 4-13% without sacrificing performance or reliability.
Access status: Open Access ,
MARLIN: multi-agent game-theoretic reinforcement learning for sustainable LLM inference in cloud datacenters
(2026-06-22) Moore, Hayden, author; Qi, Sirui, author; Milojicic, Dejan, author; Bash, Cullen, author; Pasricha, Sudeep, author; ACM, publisher
Large Language Models (LLMs) have become increasingly prevalent in cloud-based platforms, propelled by the introduction of AI-based consumer and enterprise services. LLM inference requests in particular account for up to 90% of total LLM lifecycle energy use, dwarfing training energy costs. The rising volume of LLM inference requests is increasing environmental footprints, particularly carbon emissions and water consumption. To improve sustainability for LLM inference serving in cloud datacenter environments, we propose a novel multi-agent game-theoretic reinforcement learning framework called MARLIN to co-optimize time-to-first token (TTFT), carbon emissions, water usage, and energy costs associated with LLM inference. MARLIN demonstrates a reduction of at least 18% in TTFT, 33% in carbon emissions, 43% in water usage, and 11% in energy costs compared to state-of-the-art LLM inference management frameworks.
Access status: Open Access ,
A mixed-signal implementation of an ADC-less ping-pong temporal front-end for sensor-to-photonic processing
(2026-06-22) Ariyak, Goker, author; Nikdast, Mahdi, author; ACM, publisher
Edge sensing systems require front-end architectures that reduce data movement, latency, and energy before full downstream processing. This paper presents a mixed-signal implementation and system-level adaptation of an ADC-less ping-pong temporal sensor front-end for low-overhead frame-difference sensing. The underlying ping-pong PWM pixel principle is adopted from prior CMOS vision-sensor work; this paper does not claim the base ping-pong pixel concept itself as new. Instead, the contribution is the implementation, validation, and sensor-to-photonic interface adaptation of that temporal front-end. The adopted front-end uses odd/even phased analog state reuse and in-pixel temporal differencing to extract motion-relevant information before full-frame digitization. A PWM representation of temporal difference is digitized by a compact counter-based readout path, producing a DATA_VALID-qualified digital interface. Preliminary one-pixel mixed-signal results in a TSMC 0.18 μm environment demonstrate functionally correct operation, stable analog boundary waveforms, verified odd/even lane alternation, and repeatable interleaved 8-bit code capture. Simulation-extracted readout timing shows a 1.017 μs DATA_VALID window with zero odd/even overlap, while current architecture-level estimates indicate 70 μW to 75 μW total power and 700 fps to 1000 fps potential. The same front-end provides an electrical interface to a downstream near-sensor photonic processing stage, where temporally preprocessed outputs can be converted to optical activations rather than exporting raw image data.
Access status: Open Access ,
Privacy in federated learning models for intrusion detection systems
(2026-08-06) Schneider, Erik, author; Bezawada, Bruhadeshwar, author; Ray, Indrakshi, author; ACM, publisher
Federated learning-based solutions enable multiple organizations to build a machine learning model using the collective knowledge of the organizations without exchanging the sensitive raw data of any participant. The key roadblock for widespread adoption is the possibility of training data leakage in federated learning models, which have been reported in federated learning models for image processing, natural language processing, and tabular data. However, the impact and feasibility of such privacy leakage attacks on federated learning-based intrusion detection systems (IDS) remains largely unexplored. In this work, for the first time, we focus on the problem of training data privacy in federated learning-based network IDS by exploring the implementations, techniques, and limitations of existing federated IDS solutions. First, we provide a systematic study of the landscape of machine learning-based network IDS and federated learning-based network IDS. Second, we demonstrate that data leakage attacks on federated learning-based IDS are yet to be explored to the full extent by the research community. Finally, we describe the existing defenses against such attacks and suggest the possibility of using these approaches for federated network IDS in the future.