Accurate identification of animal behaviors is fundamental to behavioral neuroscience, yet manual annotation remains a major bottleneck due to its subjectivity, limited scalability, and high labor cost. This work presents a unified deep learning and heuristic-based framework for automated behavior classification in the Novel Object Recognition Test (NORT), a widely used paradigm for assessing memory and cognitive function in rodents. We fine-tuned a YOLOv11 Pose model to detect nose and tail-base keypoints specific to Long-Evans rats, and used these keypoints to derive spatial heuristics for identifying NORT behaviors. In parallel, we trained YOLOv11 classification models to predict Standing, Object Interaction, and Other behaviors from individual frames, using datasets containing either 2-object or 5-object videos, as well as a combined dataset containing both 2-object and 5-object videos. Our model trained on the combined dataset, enhanced with heuristic-based post-processing to identify the specific objects that the rat is interacting with, achieved the best overall accuracy and generalization across arenas, outperforming heuristic-only and subset-specific 2-object or 5-object models. Detailed confusion matrices and error analyses reveal that most errors occur near behavioral transitions, reflecting the inherent ambiguity of manual labels at frame-level. The proposed framework enables scalable, reproducible, and objective annotation of NORT videos and provides a foundation for future extensions toward temporally-aware behavioral analysis. All the code and model weights, together with sample NORT videos, will be made publicly available to support future rodent behavior recognition research.

This paper presents a novel use case of Graph Convolutional Network (GCN) learning representations for predictive data mining, specifically from user/task data in the domain of high-performance computing (HPC). It outlines an approach based on a coalesced data set: logs from the Slurm workload manager, joined with user experience survey data from computational cluster users. We introduce a new method of constructing a heterogeneous unweighted HPC graph consisting of multiple typed nodes after revealing the manifold relations between the nodes. The GCN structure used here supports two tasks: i) determining whether a job will complete or fail and ii) predicting memory and CPU requirements by training the GCN semi-supervised classification model and regression models on the generated graph. The graph is partitioned into partitions using graph clustering. We conducted classification and regression experiments using the proposed framework on our HPC log dataset and evaluated predictions by our trained models against baselines using test_score, F1-score, precision, recall for classification, and R1 score for regression, showing that our framework achieves significant improvements.

Given the cost of HPC clusters, making best use of them is crucial to improve infrastructure ROI. Likewise, reducing failed HPC jobs and related waste in terms of user wait times is crucial to improve HPC user productivity (aka human ROI). While most efforts (e.g., debugging HPC programs) explore technical aspects to improve ROI of HPC clusters, we hypothesize non-technical (human) aspects are worth exploring to make non-trivial ROI gains; specifically, understanding non-technical aspects and how they contribute to the failure of HPC jobs. In this regard, we conducted a case study in the context of Beocat cluster at Kansas State University. The purpose of the study was to learn the reasons why users terminate jobs and to quantify wasted computations in such jobs in terms of system utilization and user wait time. The data from the case study helped identify interesting and actionable reasons why users terminate HPC jobs. It also helped confirm that user terminated jobs may be associated with non-trivial amount of wasted computation, which if reduced can help improve the ROI of HPC clusters.

The need for increased performance of mobile device directly conflicts with the desire for longer battery life. Offloading computation to resourceful servers is an effective method to reduce energy consumption and enhance performance for mobile applications. Today, most mobile devices have fast wireless link such as 4G and Wi-Fi, making cloud platform a good destination for computation offloading. Android provides mechanisms for creating mobile applications but lacks a native scheduling system for determining where code should be executed. This paper presents Jade, a system that adds sophisticated energy-aware computation offloading capabilities to Android applications. Jade monitors device and application status and automatically decides where code should be executed. Jade dynamically adjusts offloading strategy by adapting to workload variation, communication costs, and device status. Jade minimizes the burden on developers to build applications with computation offloading ability by providing easy-to-use Jade API. Evaluation shows that Jade can effectively reduce up to 35% of average power consumption for mobile device while improving application performance.

Scientific computational experiments often span multiple computational and analytical steps, and during execution, researchers need to store, access, transfer, and query information. Scientific workflow is a powerful tool to streamline and organize scientific application. Numbers of tools have been developed to help build scientific workflows, they provide mechanisms for creating workflow but lack a native scheduling system for determining where code should be executed. This paper presents Emerald, a system that adds sophisticated computation offloading capabilities to scientific workflows. Emerald automatically offloads computation intensive steps of scientific workflow to the cloud in order to enhance workflow performance. Emerald minimizes the burden on developers to build workflows with computation offloading ability by providing easy-to-use API. Evaluation showed that Emerald can effectively reduce up to 55% of execution time for scientific applications.

The need for increased performance of mobile device directly conflicts with the desire for longer battery life. Offloading computation to multiple devices is an effective method to reduce energy consumption and enhance performance for mobile applications. Android provides mechanisms for creating mobile applications but lacks a native scheduling system for determining where code should be executed. This paper presents Jade, a system that adds sophisticated energy-aware computation offloading capabilities to Android apps. Jade monitors device and application status and automatically decides where code should be executed. Jade dynamically adjusts offloading strategy by adapting to workload variation, communication costs, and energy status in a distributed network of Android and non-Android devices. Jade minimizes the burden on developers to build applications with computation offloading ability by providing easy-to-use Jade API. Evaluation shows that Jade can effectively reduce up to 39% of average power consumption for mobile application while improving application performance.

Mobile device users consistently desire faster results and longer battery life, which is frequently accomplished by of-floading computation to cloud-based servers. However, in common cases (Internet connectivity slow/unavailable, priva-cy/security issues), users have multiple devices with heterogeneous battery and computational capabilities independent of the cloud. Android provides mechanisms for creating mobile code, but lacks a native scheduling mechanism for determining where code should be executed. In this paper we present the results of an investigation into adding sophisticated scheduling capabilities to Android apps, which provides scheduling balancing energy and performance across networked mobile devices. Jade monitors and adapts to workload variation, communication costs, and energy status in a distributed ad-hoc network of Android mobile devices for supporting distributed computation. We show how the two goals can be integrated, and present several algorithms indicating a major advantage (over 75% improvement in energy use) can be achieved through the use of dynamic scheduling information for remote computational devices. We provide a detailed discussion of our system architecture and implementation, and briefly summarize the experimental results which have been achieved.

The Simple Object Access Protocol (SOAP) is a dominant enabling technology in the field of Web services. Web services demand high performance, security and extensibility. SOAP, being based on extensible markup language (XML), inherits not only the advantages of XML, but its relatively poor performance. This makes SOAP a poor choice for many high-performance Web services. We present a new approach to implementing the SOAP protocol using caching on the SOAP server. This approach has significant performance advantages over current approaches while maintaining complete protocol compliance. We demonstrate its practicality by implementing a demonstration system under Linux, giving speedups of over 250% in our sample applications.

A fundamental problem in today's research environment, particularly in the biological and environmental sciences, is the inability of researchers to perform even simple investigations due to the inaccessibility and essential difficulty in acquiring and utilizing the necessary data. Often the data and simulation models are available via the Internet, but through a combination of obscurity, incompatibility and inefficiency, they are essentially unusable. Our system - the Modeler's Workbench - aims to address these fundamental problems. Novel aspects of our system include the use of XML as both a resource description language and a machine-independent data transfer mechanism. We build on top of existing metacomputing tools, while providing an infrastructure based on Java wrappers around existing simulations, to provide compatibility with our infrastructure while retaining legacy code and allowing for machine-dependent optimizations. We have developed a prototype of our system, for which we present the design and experimental data. We also discuss our future plans.

The Network Weather Service (NWS) is a distributed resource monitoring and utilization prediction system, employed as an aid to scheduling jobs in a metacomputing environment. We have enhanced the Network Weather Service by providing a broader metric set, forecast trend visualization, and asynchronous notification. Three new components have been added to the NWS system. The first of these is an SNMP Sensor that may be used to collect data being tracked by SNMP agents. Secondly, a Notification Process provides asynchronous events to interested applications when metrics tracked by NWS are expected to exceed established normal ranges. Finally, an extended SNMP agent daemon has been developed to provide SNMP-enabled applications with the ability to retrieve information from NWS without the need for custom programming. The new modules help to position NWS as an integral component for performing QoS monitoring and fault prediction in an SNMP-based fault management architecture. We also demonstrate that the additional functionality adds a minimal amount of overhead to the system, and reported bandwidth utilization correlates closely with results from the invasive NWS bandwidth sensor.

In this paper we present a model for dynamically scheduling HTTP requests across clusters of servers, optimizing the use of client resources as well as the scattered server nodes. We also present a system, H-SWEB, implementing our techniques and showing experimental improvements of over 250%, which have been achieved through utilizing a global approach to scheduling requests. This is the first system to provide a hierarchical scheduling mechanism for distributed HTTP server clusters incorporating dynamic client–server task distribution and distributed data access. H-SWEB uses sophisticated scheduling techniques in monitoring and adapting to workload variation at the client and server clusters for supporting typical digital library tasks, such as fast WWW image browsing. We provide a discussion of our system architecture and implementation, and briefly summarize the experimental and analytical results that have been achieved. Copyright © 2000 John Wiley & Sons, Ltd.

A significant number of new Java-based technologies for mobile code (aka agents) have recently emerged. The 'Aglets' system, from IBM's research labs, provides an elegant mechanism for creating mobile code, but lacks a native scheduling mechanism for determining where code should be executed. We present the results of an investigation into adding sophisticated scheduling capabilities to Aglets (which we refer to as brilliant Aglets, or Baglets) from the H-SWEB project, which provides hierarchical scheduling across sets of WWW server clusters. H-SWEB uses scheduling techniques in monitoring and adapting to workload variation at distributed server clusters for supporting distributed computation. We show how the two systems can be integrated, and present several algorithms indicating a major advantage (over 350%) can be achieved through the use of dynamic scheduling information. We provide a detailed discussion of our system architecture and implementation, and briefly summarize the experimental results which have been achieved.

Recently the GEMINI Holographic Particle Image Velocimetry (HPIV) system developed in the Laser Flow Diagnostics (LFD) lab at Kansas State University has been successfully applied in volumetric 3D flow velocity measurement. Due to the 3D nature of this application, very large computation and communication requirements are imposed. An innovation algorithm, the Concise Cross Correlation (CCC), is employed in the system to extract velocity field from the hologram of the test flows. With CCC we achieved a compression ratio of 10/sup 4/ and a processing speed 1000 times faster than with traditional 3D FFT-based correlation. To further accelerate the processing speed for fully time- and space-resolved measurement, parallel processing is necessary. We present our design for a distributed system supporting this previously unparallelized application, and comment on our experiences implementing a master-slave distributed version of CCC utilizing MPI. Brief experimental results on Gigabit Ethernet and multiprocessor Pentium Xeon systems are given.

We present a model for dynamically scheduling HTTP requests across clusters of servers, optimizing the use of client resources as well as the scattered server nodes. We also present a system, H-SWEB, implementing our techniques and showing experimental improvements of over 250%, which have been achieved through utilizing a global approach to scheduling requests. This is the first system to provide a hierarchical scheduling mechanism for distributed HTTP server clusters incorporating dynamic client-server task distribution and distributed data access. H-SWEB uses sophisticated scheduling techniques in monitoring and adapting to workload variation at the client and server clusters for supporting typical digital library tasks, such as fast WWW image browsing. We provide a discussion of our system architecture and implementation, and briefly summarize the experimental results that have been achieved.

WWW-based information service has grown enormously during the last few years, and major performance bottlenecks have been caused by WWW server and Internet bandwidth inadequacies. Augmenting a server with multiprocessor support and shifting computation to client-site machines can substantially improve system response times and for some applications, it may also reduce network bandwidth requirements. In this paper, we propose adaptive scheduling techniques that optimize the use of a multiprocessor server with client resources by predicting demands of requests on I/O, CPU and network capabilities. We also provide a performance analysis under simplified assumptions for understanding the impact of system loads and network bandwidth when using our scheduling strategy. Finally we report preliminary experimental results to examine the system performance and verify the usefulness of the analytic model.