31 research outputs found

    Speeding-up graph processing on shared-memory platforms by optimizing scheduling and compute

    No full text
    Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2023-12-01The student, Azin Heidarshenas, accepted the attached license on 2021-12-01 at 14:42.The student, Azin Heidarshenas, submitted this Dissertation for approval on 2021-12-01 at 15:02.This Dissertation was approved for publication on 2021-12-02 at 15:20.DSpace SAF Submission Ingestion Package generated from Vireo submission #17337 on 2022-04-29 at 16:10:18Made available in DSpace on 2022-04-29T21:58:36Z (GMT). No. of bitstreams: 3 HEIDARSHENAS-DISSERTATION-2021.pdf: 3212586 bytes, checksum: 6a638d5593e66694a65f9c79a42e9744 (MD5) LICENSE.txt: 4214 bytes, checksum: 495fe180856cfd4d2a8f0cafc16b3f8c (MD5) PROQUEST_LICENSE.txt: 4560 bytes, checksum: bc34ec3f5e286621cc190f2aff207f83 (MD5) Previous issue date: 2021-12-02Embargo set by: Seth Robbins for item 123459 Lift date: 2024-04-29T21:58:46Z Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemAuthor requested closed access (OA after 2yrs) in Vireo ETD systemLimitedGraph processing workloads are being widely used in many domains such as computational biology, social network analysis, and financial analysis. As DRAM technology scales down into higher densities, shared-memory platforms gain increasing importance in handling large graph sizes. We study two main categories of graph algorithms from an implementation perspective. Topology-driven algorithms process all vertices of the graph at each iteration, while data-driven algorithms only process those vertices that make a substantial contribution to the output. Furthermore, the performance of a graph algorithm execution can be broken down into three components, namely, pre-processing, compute, and scheduling. For data-driven algorithms, the work of each thread is driven by the dependencies between vertex values that are known only at run-time. Hence, the scheduling will take a significant portion of execution. However, for topology-driven algorithms, the scheduling time is negligible since the work of each thread can be determined at compile-time. In this dissertation, we present three techniques to address the performance bottlenecks in both data-driven and topology-driven algorithms. First, we present Snug, which is a chip-level architecture that mitigates the trade-off between synchronization and wasted work in data-driven algorithms. Second, we present V-Combiner, which is a software-only technique to mitigate the trade-off between performance and accuracy of topology-driven algorithms using novel vertex-merging and recovery mechanisms. Finally, we present KeepCompressed, which is a set of algorithms to speed-up compute for topology-driven algorithms using vertex clustering for dynamic graphs

    Architectural support for work-efficient relaxed priority queueing

    No full text
    Limited Restriction set for Item 102680 on 2019-04-29T15:20:14Z with date 2020-08-10 by [email protected] Restriction set for Item 102680 on 2019-04-29T15:20:29Z with date 2020-08-10 by [email protected] parallel algorithms in domains such as graph analytics and simulations execute more efficiently if some parallel tasks are executed before others. To implement such priority-based task scheduling, the data structure of choice is a concurrent priority queue (PQ). Unfortunately, PQ algorithms exhibit an undesirable tradeoff. On one hand, traditional PQs always dequeue the highest-priority task, and thus fail to scale because of contention at the head of the queue. On the other hand, relaxed PQs avoid contention by dequeuing tasks that are often so far from the head that the resulting schedule misses the benefit of priority-based scheduling. This thesis proposes novel architectural support for relaxing PQs without straying far from the priority-based schedule. Our architecture, called Snug, distributes the PQ and maintains a set of Work Registers that point to the highest-priority task in each subqueue. Snug provides an instruction that picks a high-quality task to execute. The instruction periodically switches between visiting all the subqueues to get an accurate global snapshot and visiting nearby subqueues to reduce traffic. Overall, Snug dequeues highquality tasks while simultaneously avoiding hotspots and excessive network traffic. We evaluate Snug on graph analytics and event simulation applications. Snug reduces the average execution time of the applications by 1.6×, 4.9× and 3.4× compared to the state-of-the-art skip list, SprayList, and software-distributed PQs, respectively.Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2019-05-01The student, Azin Heidarshenas, accepted the attached license on 2017-04-25 at 14:25.The student, Azin Heidarshenas, submitted this Thesis for approval on 2017-04-25 at 14:26.This Thesis was approved for publication on 2017-04-26 at 09:10.DSpace SAF Submission Ingestion Package generated from Vireo submission #11021 on 2017-08-10 at 14:32:33Made available in DSpace on 2017-08-10T19:52:22Z (GMT). No. of bitstreams: 2 HEIDARSHENAS-THESIS-2017.pdf: 760946 bytes, checksum: 812461abc69e6470e20e176c03e27ef3 (MD5) LICENSE.txt: 4214 bytes, checksum: 4e5d3a69289d94ed974426c0a5808e16 (MD5) Previous issue date: 2017-04-26Embargo set by: Colleen Fallaw for item 102680 Lift date: 2019-08-10T21:25:30Z Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemLimited Restriction set for Item 102680 on 2019-04-29T15:20:40Z with date 2020-08-10 by [email protected] Restriction set for Item 102680 on 2019-04-29T15:20:43Z with date 2020-08-10 by [email protected] Restriction set for Item 102680 on 2019-04-29T15:20:46Z with date 2020-08-10 by [email protected] Restriction Lifted for Item 102680 on 2020-08-10T09:15:09Z

    Polyphony in the novel The Daughter of the Peasant, you lost the work to Azin using the approach of dialogue

    No full text
    Abstract This research applies Bakhtin’s views and ideas, especially dialogism theory, to critically analyze the discourse in the novel “The Peasant’s Daughter” by the contemporary Iranian novelist Beh Azin to reflect on the polyphony of this novel. This study answers the following questions: 1- How do women in the novel appear dynamically in polyphony? 2. How does the conflict between modernity and tradition appear in polyphony in the novel? 3. How do intellectuals and elites influence the emergence of polyphony in the novel? The results show women were passive to represent polyphony, providing a narrative context for the dominant picture of the protagonist’s discourse. The elites in the story use polyphony for various reasons because of left-wing and nationalist ideologies. The confrontation between tradition and modernity plays no small part in the emergence of this uprising, in which tradition helped the wealthy classes to institutionalize themselves in society. In conclusion, social background and gender are two prominent issues in the story’s narration of The Peasant Girl. The author challenges traditional society’s ideological and moral roots while establishing polyphony between the bourgeoisie and the subjugated classes

    RUMBA: Runtime Monitoring and Behavioral Analysis Framework for Java Software Systems

    No full text
    I hereby declare that I am the sole author of this thesis. This is a true copy of the thesis, including any required final revisions, as accepted by my examiners. I understand that my thesis may be made electronically available to the public. ii A goal of runtime monitoring is to observe software execution to determine whether it complies with its intended behavior. Monitoring allows one to analyze and recover from detected faults, providing prevention activities against catastrophic failure. Although runtime monitoring has been in use for so many years, there is renewed interest in its application largely because of the increasing complexity and ubiquitous nature of software systems. To address such a demand for runtime monitoring and behavioral analysis of software systems, we present RUMBA framework. It utilizes a synergy between static and dynamic analyses to evaluate whether a program behavior complies with specified properties dur-ing its execution. The framework is comprised of three steps, namely: i) Extractin

    Architectural support for work-efficient relaxed priority queueing

    No full text
    Many parallel algorithms in domains such as graph analytics and simulations execute more efficiently if some parallel tasks are executed before others. To implement such priority-based task scheduling, the data structure of choice is a concurrent priority queue (PQ). Unfortunately, PQ algorithms exhibit an undesirable tradeoff. On one hand, traditional PQs always dequeue the highest-priority task, and thus fail to scale because of contention at the head of the queue. On the other hand, relaxed PQs avoid contention by dequeuing tasks that are often so far from the head that the resulting schedule misses the benefit of priority-based scheduling. This thesis proposes novel architectural support for relaxing PQs without straying far from the priority-based schedule. Our architecture, called Snug, distributes the PQ and maintains a set of Work Registers that point to the highest-priority task in each subqueue. Snug provides an instruction that picks a high-quality task to execute. The instruction periodically switches between visiting all the subqueues to get an accurate global snapshot and visiting nearby subqueues to reduce traffic. Overall, Snug dequeues highquality tasks while simultaneously avoiding hotspots and excessive network traffic. We evaluate Snug on graph analytics and event simulation applications. Snug reduces the average execution time of the applications by 1.6×, 4.9× and 3.4× compared to the state-of-the-art skip list, SprayList, and software-distributed PQs, respectively.LimitedAuthor requested closed access (OA after 2yrs) in Vireo ETD system. Author then requested an extension of closed access for 1 year

    The Persian ney: A study of the instrument and its musical style

    No full text
    In view of the fact that the Ney-e Haftband is the only wind tradition in classical Persian music, the lack of attention paid to it in previous scholarly studies is rather surprising. This dissertation is a study of the Persian ney and its music as it emerged in the city of Esfehan during the nineteenth century.It provides a comprehensive body of information related to the structure of the instrument, its playing techniques, and its use in current musical practices. It discusses the development of the Persian technique by Nayeb Assdollah, the famous ney player of the Nasseri court, and how his new technique of sound production, the 'tooth and lip' technique, provided possibilities for development and change in the musical style of the ney. An extended section discusses the role of Ostad Hassan Kassai in the dissemination of current musical practices on the ney. Kassai's ideas and concepts about poetry and musical pulse have established definite styles in the tradition of ney playing and have resulted in the incorporation of many elements from both vocal music and the music of string instruments.The second section of the study includes an analytical discussion of 17 performances of classical Persian music by four of Iran's current ney players. It also includes a few performances by some former musicians in order to cover earlier trends in the practice of the ney. The performances are from the dastgahs of Shur, Segah, Mahur, and Chahargah. The analyses cover the choice of gushes and how these choices are affected by their adaptability to the specific characteristics of the ney. The dominant and principal gushes of each performance are transcribed in an attempt to understand how the musical content of each gushe compares or contrasts to its original form in radif books.The study has found that the structure of the ney does determine to a large extent, the process of improvisation and the selection of gushes in performances. However, despite substantial variety, improvisations on the ney are found to be compatible with the practice of radif on other instruments, and that previous assumptions regarding an autonomy in the radif of ney is not valid. Moreover, the use of gushes and the specific roles assigned to them coincide, for the most part, with their traditional use in radif books. In conclusion, the study has determined that the music of the ney belongs to a musical system which stresses individual traditions, but at the same time is unified by closely related structures and symbols.Made available in DSpace on 2011-05-07T12:59:09Z (GMT). No. of bitstreams: 2 license.txt: 4922 bytes, checksum: 910b249b4beec47e7ab768910c8f966f (MD5) 9329115.pdf: 8500112 bytes, checksum: 4a668d99be4b2ae8ef1d3ff1f548f797 (MD5) Previous issue date: 1993Item marked as restricted to the 'UIUC Users [automated]' Group (id=2) by Howard Ding ([email protected]) on 2011-05-07T14:48:40Z Item is restricted indefinitely.Restriction data tranferred 2014-07-01T11:22:03-05:00 Original Data Group with Access UIUC Users [automated] Release Date: none Reason: ETDs are only available to UIUC Users without author permissionETDs are only available to UIUC Users without author permissionU of I Onl

    Evaluating Extracorporeal Membrane Oxygenation and Ventilation Treatment of Patients with COVID-19: A Review

    No full text
    The prevalence of COVID-19 in the world is rapidly increasing. Although some patients show mild symptoms of the virus, some others need special care due to the exacerbation of the disease. Therefore, invasive treatments are needed to treat these patients. Data were collected from PubMed and Google scholars at various time points up to the 2020 academic year. The related keywords are listed as follows: "COVID-19", "Treatment", "Pathogenesis", and "Lung disorder". Studies have shown that although the use of ECMO and ventilation can provide oxygen to patients and improve their clinical status; these procedures can lead to the activation of inflammatory responses and the activation of the renin-angiotensin system. Inflammation and activation of the renin-angiotensin system are among the weak prognoses for COVID-19-infected patients. ECMO and ventilation treatment procedures are like double-edged swords, and monitoring patients during treatment is essential to prevent renin-angiotensin activation. *Corresponding Author: Zahra Gatavizadeh; Email: [email protected] Please cite this article as: Hybar H, Samimi A, Maniati M, Jodat J, Gatavizadeh Z. Evaluating Extracorporeal Membrane Oxygenation and Ventilation Treatment of Patients with COVID-19: A Review. Arch Med Lab Sci. 2020;6:1-6 (e16). https://doi.org/10.22037/amls.v6.3355

    Positioning of anchors for personal fall arrest systems for sloped roofs

    No full text
    2019 Spring.Includes bibliographical references.Construction worker falls account for about one-third of all construction fatalities increasing 53% from 2011 to 2015 with most fatalities in the roofing trade. A personal fall arrest system (PFAS) is an effective means of fall protection and required by the Occupational Safety and Health Administration (OSHA). However, PFAS anchor point placement is an issue evidenced by the number of fatalities caused by incorrect anchor positioning (Hinze and Olbina, 2008). This research looks at a process to create a tool to optimize PFAS anchor points by: 1) Converting OSHA and American National Standards Institute (ANSI) regulations and standards for anchor point positions into computer-readable format; and 2) Developing a tool for optimization of locations of anchor points. A qualified field user performs the tool data input. Data include PFAS features (e.g. lanyard length) and project-related values (e.g. roof height). The tool then looks for the potential anchor locations that satisfy the fall clearance and swing hazard requirements. K-Nearest Neighbor Search (KNNS) algorithm was used as the optimization model for the tool. The tool was developed in Python programming language and was compiled into a standalone computer application with a user-friendly interface. The output of the tool includes optimized anchor points displayed both graphically and numerically. The tool results were validated using the K-fold Cross-Validation method and proved the tool output results to be adequately accurate. The contribution of this research is the development of an automated field-level process for steep sloped roofing companies that would help improve their safety practices

    Overheating assessment for apartments during average and hot summers in the Nordic climate

    No full text
    Publisher Copyright: © 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.Amidst rising ambient temperatures and frequent heatwaves, assessing summer indoor overheating is crucial. This study examined overheating in over 6000 apartments to show the temporal aspects of indoor overheating during the hot summer of 2021 compared to the average summer of 2020 in the Helsinki region of Finland, with additional consideration of the apartments’ design year, size, and number of rooms. Data analysis was conducted employing ANOVA and post hoc tests to assess the impact of apartment characteristics on overheating. Finnish building codes and health-related regulations defined three temperature thresholds to determine overheating: 27°C, 30°C, and 32°C. The findings indicated that during the summer of 2020, 76%, 8%, and 1% of apartments exceeded 27°C, 30°C, and 32°C, respectively. In the hot summer of 2021, these proportions increased significantly to 96%, 32%, and 4%. The average number of degree hours above 27°C in 2021 was three times higher than in 2020. While nighttime temperatures were slightly lower, there was a considerable number of degree hours above 27°C. Smaller apartments exhibited a higher risk of overheating. Those constructed after 2012 demonstrated lower overheating risks during the hot summer.Peer reviewe
    corecore