Search results

Title: Gospel Mission Baptist Church, Chicago, Illinois, ca. 1945
Creator: Illinois Institute of Technology
Date: 1943-1948
Description: Photograph of the front of the Gospel Mission Baptist Church. Date of photograph is unknown. Date range listed is approximate.
Collection: Office of Communications and Marketing photographs, 1905-1999

Title: Federal Arts Project mosaic, Monterey, California
Description: Photograph of a mural completed in Monterey, California as part of a the Federal Arts Project by Mary Dill, likely in the late 1930s,...
Show morePhotograph of a mural completed in Monterey, California as part of a the Federal Arts Project by Mary Dill, likely in the late 1930s, described on the inscription of a similar photograph as "Mary Dill's first mosaic for the Federal Art Project - the first governor's house in Monterey, Calif." Photographer unknown. Date of photograph unknown.
Show less
Collection: Mary Dill Henry Papers, 1913-2021

Title: Federal Arts Project mosaic, Monterey, California
Description: Photograph of a mural completed in Monterey, California as part of a the Federal Arts Project by Mary Dill, likely in the late 1930s,...
Show morePhotograph of a mural completed in Monterey, California as part of a the Federal Arts Project by Mary Dill, likely in the late 1930s, described on the inscription of a similar photograph as "Mary Dill's first mosaic for the Federal Art Project - the first governor's house in Monterey, Calif." Photographer unknown. Date of photograph unknown.
Show less
Collection: Mary Dill Henry Papers, 1913-2021

Title: Federal Arts Project mosaic, Monterey, California
Description: Photograph of a mural completed in Monterey, California as part of a the Federal Arts Project by Mary Dill, likely in the late 1930s,...
Show morePhotograph of a mural completed in Monterey, California as part of a the Federal Arts Project by Mary Dill, likely in the late 1930s, described on the inscription of a similar photograph as "Mary Dill's first mosaic for the Federal Art Project - the first governor's house in Monterey, Calif." Photographer unknown. Date of photograph unknown.
Show less
Collection: Mary Dill Henry Papers, 1913-2021

Title: Federal Arts Project mosaic, Monterey, California
Description: Photograph of a mural completed in Monterey, California as part of a the Federal Arts Project by Mary Dill, likely in the late 1930s....
Show morePhotograph of a mural completed in Monterey, California as part of a the Federal Arts Project by Mary Dill, likely in the late 1930s. Inscription on verso: "Mary Dill's first mosaic for the Federal Art Project - the first governor's house in Monterey, Calif." Photographer unknown. Date of photograph unknown.
Show less
Collection: Mary Dill Henry Papers, 1913-2021

Title: An architectural and structural study of multi-storied skeleton building
Creator: Lempp Kerbis, Gertrude 1926-2016
Date: 1954

Title: LOCAL VISCOELASTIC PROPERTIES OF SOFT ANISOTROPIC FIBROUS TISSUE
Creator: Gallo, Nicolas Remy
Date: 2020
Description: The current aging population, with more than 80 million "baby boomers", will present a steep medical challenge for our society in a...
Show moreThe current aging population, with more than 80 million "baby boomers", will present a steep medical challenge for our society in a foreseeable future. Half of the adults over 85 years old are predicted to be diagnosed with Alzheimer's disease by 2050. With healthcare cost reaching over 700 billion dollars in the United States, early detection of Alzheimer's disease (AD) and other co-existing neurodegenerative diseases is crucial to improve the recovery odds in patients and to decrease individual care cost. This work seeks to tackle this problem by proposing a novel computational framework toward improving the measurement of shear visco-elastic properties of brain white matter (WM), which vary with age. These measurements practically represent the effective (average) response of many cells and are typically obtained by using rheology or elastography. Although the former is direct, the latter requires the solution of an inverse problem based on a priori mechanical tissue model. The mechanical anisotropy of WM has previously not been fully explored although many inconsistencies have been reported in brain MRE experiments. To account for these inconsistencies a transversely isotropic constitutive model for the brain WM is proposed to interpret prior experiments involving 7 young and 4 older healthy men. By employing a novel inversion scheme, we report the local variation of the effective transverse and axial shear moduli in two well aligned WM structures (corpus callosum: CC; and cortical spinal tract: CST) for both the young and old cohort of healthy subjects part of the study. This work reports statistically significant changes in local regional variation of the transverse modulus across the CC for the young cohort. In the older cohort, the trend was similar yet not statistically significant. A novel candidate biomarker, the shear anisotropy metric, defined as the ratio of the transverse and axial shear moduli, found statistically significant local regional variation across the CC but not in the CST. Healthy aging was observed to decrease both transverse and axial in both CC and CST, although the variation was significant only for the CC. Finally, in an effort to understand the cause of effective transverse mechanical properties variation in WM with aging, the connection between effective and intrinsic contribution of WM cellular constituents is established. The intrinsic mechanical contributions of axons and glial matrix are separated by fitting the estimates of the effective shear moduli to a microscopic composite fiber model of myelinated axons embedded in the glial matrix. This work provides a method to establish a baseline for healthy brain mechanical properties thus promising to increase the specificity of MRE toward early diagnosis of neurodegenerative diseases. Additional oscillating disc rheology experiments with decellularized porcine myocardium, and the fabrication of a stable heterogeneous phantom matching the mechanical, diffusional and electrical properties of the WM provide foundational knowledge for due development and validation of MRE methodologies employed in other tissues.
Show less

Title: Students building a snowman, Illinois Institute of Technology, Chicago, Illinois, 1980s
Creator: Lightfoot, Robert M., III
Date: 1980-1989
Description: Photograph of students building a snowman on present day Siegel Field, north of S.R. Crown Hall on the Illinois Tech campus. Date of...
Show morePhotograph of students building a snowman on present day Siegel Field, north of S.R. Crown Hall on the Illinois Tech campus. Date of photograph is unknown. Date range listed is approximate.
Show less
Collection: Office of Communications and Marketing photographs, 1905-1999

Title: Spring Thing tricycle race, Illinois Institute of Technology, Chicago, Illinois, ca. 1968-1971
Date: 1968-1971
Description: Photograph of the tricycle race during the Spring Thing. Spring Thing, sponsored by the Union Board, occurred during the fall semester,...
Show morePhotograph of the tricycle race during the Spring Thing. Spring Thing, sponsored by the Union Board, occurred during the fall semester, usually in October. The tricycle race, first held in 1968, was a highlight of the annual festivities. Photographer unknown. Date of photograph is unknown. Date range listed is approximate.
Show less
Collection: Office of Communications and Marketing photographs, 1905-1999

Title: Women's Hall (later Lewis Hall), Illinois Institute of Technology, Chicago, Illinois, ca. 1966
Date: 1965-1966
Description: Photograph of Lewis Hall, designed by Mittelbusher & Tourtelot and constructed in 1966. The building is located at 70 East 33rd Street, and...
Show morePhotograph of Lewis Hall, designed by Mittelbusher & Tourtelot and constructed in 1966. The building is located at 70 East 33rd Street, and was known at the time of its construction as Women's Hall. Date of photograph is unknown. Date range listed is approximate. Photographer unknown.
Show less
Collection: Office of Communications and Marketing photographs, 1905-1999

Title: My IIT
Creator: Namburi, Kartheeka
Date: 2024
Description: Master's Project focused on developing a user-friendly mobile application for Illinois Institute of Technology (IIT).

Title: Laser Powder Bed Fusion Of Cost-Effective Non-Spherical Ti-6Al-4V Powder
Creator: Asherloo, Mohammadreza
Date: 2023
Description: This comprehensive research delves into the intricate dynamics of Laser Powder Bed Fusion (L-PBF) of Ti-6Al-4V powders, emphasizing the...
Show moreThis comprehensive research delves into the intricate dynamics of Laser Powder Bed Fusion (L-PBF) of Ti-6Al-4V powders, emphasizing the potential of non-spherical, hydride-dehydride (HDH) powders as a cost-efficient alternative to traditional spherical powders. The study systematically explores the interplay between powder morphology, granulometry, and various post-processing treatments in shaping the resultant microstructure, porosity, and mechanical properties of L-PBF fabricated Ti-6Al-4V components.Initial investigations focused on the flowability, packing density, and resultant density of L-PBF parts using HDH powders with varying size distributions. Through meticulous optimization of laser parameters, parts with a relative density exceeding 99.5% were achieved, even at production rates 1.5–2 times higher than conventional LPBF processes. Dynamic synchrotron X-ray imaging provided insights into laser-powder interactions, revealing key mechanisms of porosity formation associated with HDH powders. Further microstructural examinations highlighted the formation of columnar β grains with acicular α/α′ phases in the as-built condition. Mechanical tests, including fatigue assessments under fully-reversed tension-compression conditions, revealed the critical role of surface roughness in fatigue performance. Notably, mechanical grinding significantly improved fatigue strength, especially in the high cycle fatigue region, by eliminating surface micro-notches. X-ray diffraction analyses further elucidated the stress and micro-strain profiles, offering insights into the material's deformation mechanisms. A pivotal discovery was the presence of α/α′ on prior β/β grain boundaries, challenging the prevailing notion that high cooling rates in L-PBF preclude β/β grain boundary variant selection. Electron backscatter diffraction and synchrotron X-ray imaging illuminated the role of powder characteristics in locally modulating cooling rates, leading to β/β grain boundary α′ lath growth. Lastly, the research underscored the multifaceted interdependencies among contouring, powder granulometry, Hot Isostatic Pressing (HIP), and mechanical surface treatments. A pronounced increase in sub-surface porosities was identified when contouring was combined with fine powder granulometry. However, post-HIP treatments induced a phase transformation from martensitic α′ to a basket-weave α+β microstructure, enhancing the material's fatigue resistance to levels comparable to wrought Ti-6Al-4V. In summation, this doctoral research offers a holistic understanding of the L-PBF process for Ti-6Al-4V, emphasizing the viability of non-spherical HDH powders and providing a roadmap for parameter optimization, defect minimization, and mechanical property enhancement in L-PBF-fabricated Ti-6Al-4V structures.
Show less

Title: Case Study: A Comparison of Pedagogical Content Knowledge Between Coaches and Coaches/Mentees
Creator: Barone, Ana MargaritaSalinas
Date: 2024
Description: This multiple case study dissertation aimed to examine one of the domains of pedagogical content knowledge, knowledge of content and students,...
Show moreThis multiple case study dissertation aimed to examine one of the domains of pedagogical content knowledge, knowledge of content and students, between different types of elementary coaches and between coach and their respective collaborating teachers. It also investigated the impact a coaches’ background experiences have on the dynamic between coaches and teachers and the perceptions' teacher have on the effectiveness of coaching. The theoretical framework used in this qualitative study was Ball, Thames, and Phelps’ (2008) definition of PCK. Data was collected from six coaches–four instructional coaches and two math coaches–and eleven k-5th grade teachers. Data collection involved a survey, LMT assessment, and semi-structured interviews, and a thematic analysis method was conducted. The findings from the cross-case analysis resulted in ten themes, with the majority having multiple categories. One finding to one of the research questions was that there were no differences in knowledge of content and students between mathematics coaches and general instructional coaches, but other areas to further investigate emerged. Another finding was that coaches were either within the same capacity as their respective teachers or had extra knowledge of content and students. Although the majority of the coaches’ knowledge of content and students was at a higher level according to their LMT score, it does not necessarily mean that coaches are working with teachers in improving knowledge of content and students. In addition, more research is recommended in creating a pedagogical content knowledge instrument that is specific for coaches.
Show less

Title: Characterization of Radiation Damage Effects in High-Energy Neutrino Target Graphite using Low-Energy Ions
Creator: Burleigh, Abraham C.
Date: 2023
Description: Exposure of graphite targets to high intensity proton beams at neutrino production facilities causes changes in the target material that can...
Show moreExposure of graphite targets to high intensity proton beams at neutrino production facilities causes changes in the target material that can result in a shortened operation lifetime. The dominant factors in this process are currently thought to be mechanical in nature resulting primarily from microstructural effects that lead to thermal and structural changes in bulk material properties. As currently planned beam facilities with increased proton energy and intensity begin to come online it will be important to thoroughly understand these processes, and ideally to be able to predict the effects of new beam designs on target properties. Direct analysis of targets exposed to existing high-energy proton beams is complicated by several factors, such as very limited access to proton beam facilities, high associated costs, irradiation times on the order of months, and the resulting radioactivity of irradiated samples that requires special facilities for post-irradiation examination. Much of the existing literature concerning irradiation damage in graphite has been focused on the needs of the nuclear engineering community, however high-energy proton targets operate in a much different environment. In comparison to graphite irradiated in a nuclear reactor, graphite used in proton beam targets receives a higher dose rate, have greater gas production, and experience short irradiation pulses as opposed to continuous irradiation. Low-energy ion irradiation offers a method of inducing similar levels of radiation damage to high-energy protons while avoiding many of the difficulties and limitations associated with high-energy proton beams and the corresponding activated specimen testing. My research described in this thesis focused on investigating how low-energy ion irradiation could be used to induce the same or similar types of microstructural alteration and mechanical property degradation as that seen in high-energy neutrino production target graphites by varying damage levels and irradiation temperatures prior to post-irradiation characterization.
Show less

Title: Nanopore sensing for environmental and biomarker analysis
Creator: Arora, Pearl
Date: 2024
Description: Nanopore stochastic sensing is a powerful analytical tool for detecting target molecules through a nanoscale pore. The analyte and electrolyte...
Show moreNanopore stochastic sensing is a powerful analytical tool for detecting target molecules through a nanoscale pore. The analyte and electrolyte ions are subjected to a voltage bias which drives them to translocate through the nanopore, resulting in disruptions in the ionic current. These disruptions are translated to blockage events which can serve as a signature of the analyte. Owing to its unique features of single-molecule and label-free sensing, nanopore technique has been exploited in a wide array of applications such as detection of metal ions, proteins, DNA, microRNA, toxic agents etc. In this dissertation, projects showcasing nanopore’s sensing capability of different biomarkers and in the detection of a wide range of target molecules based on non-covalent interactions are presented. Particularly in the first two projects, nanopore detection of ferric ions relevant to environmental regulation as well as a biomarker for human health and a miRNA-based biomarker for oral cancer and oral related diseases are summarized. Ferric ions, which are benign if present in balanced quantities but can be toxic otherwise, are detected by using an engineered multifunctional nanopore and a chelating organophosphonic acid ligand. The chelate complex formed after ferric ions bind to ligand gives significantly different event signatures than the free ligand in the solution enabling ferric ion detection. Even in the presence of interfering ions, the ferric ions could be recognized easily because of the conformational changes brought in the nanopore lumen by the interaction of the interfering metal ions with the His-tags of the nanopore which in turn resulted in variations in the characteristics of blocking events. In the second project, miR31, an oral cancer biomarker, is selectively detected with the help of an engineered nanopore, and a DNA based probe. Several probes with variations in length, composition and position of the overhangs or probes with no overhangs were compared and studied as the probes play a crucial role in capturing the target of interest with high specificity. Our strategically designed probe emerged as the most effective in capturing the target even in presence of large background from human saliva samples and enhanced the sensitivity of the system. In the first two projects, nanopores are utilized for selective and specific detection of certain target molecules. However, in order to analyze diverse range of analytes, numerous sensing systems have to be constructed which can be a time-consuming and challenging task. To circumvent this limitation, in the third project, diverse recognition sites based on various non-covalent interactions are incorporated into the α-hemolysin protein pore to achieve detection of not just a single analyte but broad category of molecules such as cations, anions, aromatic and hydrophobic compounds.
Show less

Title: Investigation of Electrochemical Properties and Fabrication of Lithium- and Sodium-ion Batteries
Creator: Chen, Changlong
Date: 2023
Description: Since the successful commercialization of Li-ion battery, the opportunity in creating a sustainable world with evenly-distributed energy...
Show moreSince the successful commercialization of Li-ion battery, the opportunity in creating a sustainable world with evenly-distributed energy supply and less environmental concerns has been significantly increased. This triggered tremendous efforts from both academy and industry in building better Li-ion batteries. Along the research and development over past 30 years, the performance of current Li-ion batteries has met some basic needs in our daily life, such as powering electronic devices and electric vehicles for a short time, while superior capabilities, like extended operating life, stable function under extreme circumstances, is always pursued. Under the pressure from these ever-growing demands, the corresponding Li-ion battery production is faced with a lot of new challenges. Regarding the battery production, the present Li-ion battery manufacturing heavily relies on the use of certain repo-toxic solvent, N-methyl-2-pyrrolidone (NMP), which arouses safety concerns to human health. In the pursuit of a higher energy density, silicon anode, bearing ten times the gravimetric capacity of commercially-dominating graphite anode, is intensively studied as the anode material for next-generation Li-ion batteries. However, its degradation mechanism is not completely revealed yet, which makes the methods of effective optimizations hard to be developed. In terms of the cost control, Na-ion batteries have been revisited and have received extra attention in the past decade owing to the abundance in raw materials and the high compatibility with state-of-art Li-ion industry while blank space in understanding primary electrochemical properties, such as impedance signals, has not been totally filled. This will also cause the misunderstandings in such interpretation and, thereby, postpone the pace of relevant advancement. Targeting these proposed issues, this thesis provides a series of feasible solutions via careful investigation and rational analysis with the aid of various advanced (non)electrochemical techniques, which offers a few unique perspectives in studying Li- and Na-ion batteries, and further facilitates the following research and development in the corresponding communities.
Show less

Title: Adaptive Learning Approach of a Domain-Aware CNN-Based Model Observer
Creator: Bogdanovic, Nebojsa
Date: 2023
Description: Application of convolutional neural networks (CNNs) for performing defect detection tasks and their use as model observers (MO) has become...
Show moreApplication of convolutional neural networks (CNNs) for performing defect detection tasks and their use as model observers (MO) has become increasingly popular in the medical imaging field. Building upon this use of CNN MOs, we have trained the CNNs to discern between the data it was trained on, and the previously unseen images. We termed this ability domain awareness. To achieve domain awareness, we are simultaneously training a new variation of U-Net CNN to perform defect detection task, as well as to reconstruct a noisy input image. We have shown that the values of the reconstruction mean squared error can be used as a good indicator of how well the algorithm performs in the defect localization task, making a big step towards developing a domain aware CNN MO. Additionally, we have proposed an adaptive learning approach for training these algorithms, and compared them to the non-adaptive learning approach. The main results that we achieved were for the ideal observers, but we also extended these results to human observer data. We have compared different architectures of CNNs with different numbers and sizes of layers, as well as introduced data augmentation to further improve upon our results. Finally, our results show that the proposed adaptive learning approach with introduced data augmentation drastically improves upon the results of a non-adaptive approach in both human and ideal observer cases.
Show less

Title: Evaluating antimicrobial efficacy of GS-2 on reusable food packaging materials
Creator: Birje, Nupoor Prasad
Date: 2024
Description: Packaging plays an important role in maintaining the quality and safety of fresh produce throughout storage, transportation and end-use by...
Show morePackaging plays an important role in maintaining the quality and safety of fresh produce throughout storage, transportation and end-use by consumers. Single-use packaging poses several environmental impacts; therefore use of reusable packaging is being encouraged in the fresh produce supply chain. However, the utilization of harmful chemicals and inadequate sanitation standards limit the reuse of packaging materials. To overcome these limitations, this study focuses on testing a non-toxic, water-soluble antimicrobial; GS-2 coating to facilitate the reuse of food packaging and reduce the risk of microbial contamination. In this study, the antimicrobial activity of GS-2 was evaluated against foodborne pathogens; Escherichia coli, Listeria monocytogenes and Salmonella enterica on plastic and cardboard coupons at 1 h and 15 min treatment times and 0.3%, 1% and 3% concentration. These coupons were also stored at 4℃ and 90% R.H. and 18℃ and 45% R.H. inoculated on different days up to 42 d with E. coli or L. monocytogenes to study retention of activity of GS-2. Additionally, the efficacy of GS-2 to reduce transfer of bacteria from cardboard and plastic to tomato was investigated. The initial level of inoculum was 9 log CFU/surface for all experiments. Cardboard and plastic without GS-2 were used to compare the reduction of bacteria on the treated surfaces. The differences in the population of bacteria were evaluated using Student’s T-Test and ANOVA; p <0.05 was considered significant. With 3% GS-2 concentration on plastic, there was > 4.50 log CFU/surface reduction of all three bacteria in 1 h. There was a lower reduction of the population on cardboard as compared to plastic for all bacteria, the reduction obtained was 1.83, 2.65 and 3.42 log CFU/surface for E. coli, L. monocytogenes and S. enterica, respectively, in 1 h. There was no significant difference between 15 min and 1 h treatments for cardboard. Further, the highest reduction of bacteria was obtained with 3% GS-2 on plastic. For cardboard, no significant difference in population reduction was obtained for E. coli or S. enterica, with 1% or 3% GS-2. However, for L. monocytogenes there was a higher reduction with 3%. GS-2 remained active on the surface of plastic and cardboard for a period of six weeks. For cardboard, there was a lower reduction of bacteria and there was no trend in the population reduction from 0 to 42 d, with the populations remaining within a range of 4-5 log CFU/surface. There was a significant transfer of E. coli or L. monocytogenes from plastic surfaces without GS-2 to tomato at 5-6 log CFU/tomato. However, the transfer of bacteria from the GS-2-coated plastic to the tomato was below the limit of enumeration. For cardboard, the population was below the limit of enumeration, irrespective of the GS-2 coating. Based on the results, GS-2 is a promising antimicrobial that reduces the microbial load on packaging surfaces and prevents cross-contamination of fresh produce. The retention of GS-2 activity makes it suitable for reusable packaging applications.
Show less

Title: Quantification of Imaging Markers at Different MRI Contrast Weightings, Vasculature, and Across Field Strengths
Creator: Nguyen, Vivian S.
Date: 2024
Description: Quantitative MRI measures physical characteristics of tissue, which creates a set scale with units that allows longitudinal monitoring and...
Show moreQuantitative MRI measures physical characteristics of tissue, which creates a set scale with units that allows longitudinal monitoring and cross-patient and cross-center studies. It enables earlier detection of disease, complements biopsy, and provides a clear numeric scale for differentiation of disease states. However, quantitative MRI acquisitions and post-processing are not trivial, which makes it hard to implement the clinical setting. This along with the variability in clinically used acquisitions and post-processing techniques leads to difficulty in establishing reliable, consistent, and accurate quantitative information. There is a critical need for rigorous validation of quantitative imaging biomarkers, both for current and novel quantitative imaging techniques. This dissertation seeks to both validate current quantitative MR imaging techniques and develop new ones in the heart and brain by: 1) examining the data variability and the loss in tag fidelity that occurs when quantitative cardiac tagging is incorrectly run post-Gadolinium injection; 2) quantifying the negative impact of unexpected relaxometric behavior observed in low field MR imaging for low inversion times during T1 mapping; 3) validating retrospectively calculated T1 as a biomarker for Multiple Sclerosis progression; 4) and prototyping an oxygen extraction fraction (OEF) mapping technique for the purpose of stroke prediction and establishment of a numeric scale for tissue health for stroke patients.Assessment of pre-Gadolinium and post-Gadolinium cardiac tag quality showed that post-Gadolinium tags are less saturated (p = 0.012) and have a wider range of saturation, contrast, and sharpness. This results in a loss of information in the late cardiac cycle and impeding quantification of myocardial function.Investigation of 64mT T1 mapping revealed unique relaxometric behavior in that at low inversion times (<250 ms), the signal response curve displayed an increase in signal intensity or a plateau in signal intensity dependent on T1 relaxation time. Inclusion of this increase or plateau in signal intensity negatively impacted T1 fitting algorithms, leading to their failure or incorrectly calculated T1 values. The maximum peak signal intensity before the null point was found to be 210 ms, which impacts current low field T1 mapping protocols which use an initial inversion time of 80-110 ms.Validation of retrospectively calculated T1 as a biomarker in Multiple Sclerosis revealed that T1 of normal appearing brain tissue correlates with measures of Multiple Sclerosis progression (EDSS, BPF, and disease duration) with normal appearing white matter T1 correlating with BPF (r = -0.49, p = 0.0018); putamen T1 correlating with EDSS (r = 0.48, p = 2.40e-03), with BPF (r = 0.69, p = 2.04e-06), and disease duration (r = -0.37; p = 0.02); and globus pallidus T1 correlating with disease duration (r = -0.42; p = 0.0093). Lesion T1 is reflective of MS severity whereas MTR is not.Finally, development of an oxygen extraction fraction (OEF) mapping technique showed that application of independent component analysis (ICA) to cardiac gated spiral-trajectory phase images yielded components that feature stenosis features observed in magnitude images. These ICA components form the basis of OEF mapping from phase images. This dissertation presents four studies that seek to improve either current quantitative MR imaging protocols in the heart, or to develop and validate new quantitative MR imaging techniques in the brain for the purpose of monitoring disease progression or predicting disease.
Show less

Title: Utilizing Concurrent Data Accesses for Data-Driven and AI Applications
Creator: Lu, Xiaoyang
Date: 2024
Description: In the evolving landscape of data-driven and AI applications, the imperative for reducing data access delay has never been more critical,...
Show moreIn the evolving landscape of data-driven and AI applications, the imperative for reducing data access delay has never been more critical, especially as these applications increasingly underpin modern daily life. Traditionally, architectural optimizations in computing systems have concentrated on data locality, utilizing temporal and spatial locality to enhance data access performance by maximizing data and data block reuse. However, as poor locality is a common characteristic of data-driven and AI applications, utilizing data access concurrency emerges as a promising avenue to optimize the performance of evolving data-driven and AI application workloads.This dissertation advocates utilizing concurrent data accesses to enhance performance in data-driven and AI applications, addressing a significant research gap in the integration of data concurrency for performance improvement. It introduces a suite of innovative case studies, including a prefetching framework that dynamically adjusts aggressiveness based on data concurrency, a cache partitioning framework that balances application demands with concurrency, a concurrency-aware cache management framework to reduce costly cache misses, a holistic cache management framework that considers both data locality and concurrency to fine-tune decisions, and an accelerator design for sparse matrix multiplication that optimizes adaptive execution flow and incorporates concurrency-aware cache optimizations.Our comprehensive evaluations demonstrate that the implemented concurrency-aware frameworks significantly enhance the performance of data-driven and AI applications by leveraging data access concurrency.Specifically, our prefetch framework boosts performance by 17.3%, our cache partitioning framework surpasses locality-based approaches by 15.5%, and our cache management framework achieves a 10.3% performance increase over prior works. Furthermore, our holistic cache management framework enhances performance further, achieving a 13.7% speedup. Additionally, our sparse matrix multiplication accelerator outperforms existing accelerators by a factor of 2.1.As optimizing data locality in data-driven and AI applications becomes increasingly challenging, this dissertation demonstrates that utilizing concurrency can still yield significant performance enhancements, offering new insights and actionable examples for the field. This dissertation not only bridges the identified research gap but also establishes a foundation for further exploration of the full potential of concurrency in data-driven and AI applications and architectures, aiming at fulfilling the evolving performance demands of modern and future computing systems.
Show less

repository.iit

Search the repository

Pages

Pages

Enabled Filters

Refine Results

Type

Date

Academic status of creator

Department

Subject

Creator

Part of

Rights