Browsing by Author "Liu, Feng"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
Item Elastic Job Bundling: An Adaptive Resource Request Strategy for Large-Scale Parallel Applications(2015-04-16) Liu, Feng; Weissman, JonIn today’s batch queue HPC cluster systems, the user submits a job requesting a fixed number of processors. The system will not start the job until all of the requested resources become available simultaneously. When cluster workload is high, large sized jobs will experience long waiting time due to this policy. In this paper, we propose a new approach that dynamically decomposes a large job into smaller ones to reduce waiting time, and lets the application expand across multiple subjobs while continuously achieving progress. This approach has three bene?ts: (i) application turnaround time is reduced, (ii) system fragmentation is diminished, and (iii) fairness is promoted. Our approach does not depend on job queue time prediction but exploits available back?ll opportunities. Simulation results have shown that our approach can reduce application mean turnaround time by up to 48%.Item Elastic Scheduling in HPC Resource Management Systems(2018-12) Liu, FengHigh Performance Computing (HPC) aggregates the power of computer clusters to tackle large problems empowering science. HPC resource scheduling today is faced with multiple challenges. Firstly, most HPC clusters are managed by queue batch systems. Batch scheduler maximizes application run-time efficiency while sacrifices response time and sometimes utilization. Secondly, HPC clusters reserved for on-demand data analysis are operated at low utilization. Thirdly, multiple heterogeneous and dynamic HPC resources greatly complicate resource scheduling for distributed applications. To solve these problems, this thesis presents several elastic scheduling approaches. Elasticity means the ability to dynamically allocate resources based on workloads. Elasticity is commonly supported in Cloud but is lacking in HPC. Our approaches include new scheduling algorithms and implementations of the algorithms as services. Our services leverage existing techniques and are non-invasive, meaning that they minimize the changes to user interfaces. We address the first problem using Elastic Job Bundling (EJB), a technique that dynamically transforms a large batch job into multiple smaller subjobs so that the subjobs will start early on immediately available resources. Simulation results show that our approach reduces application mean turnaround time by up to 48%, reduces resource fragmentation by up to 59%, and reduces priority inversions by 20%. We address the second problem using Balancer, a technique that combines and dynamically moves nodes between an on-demand cluster and a batch cluster. Our results show that for a real-life scenario, our approach reduces the current investment in on-demand cluster by 82% while at the same time improving the mean batch wait time by 8x. We address the third problem using Bundle, a resource abstraction that represents heterogeneous resource capacities and capabilities in a uniform way. We implement Bundle as a service on 10+ heterogeneous HPC resources. We use Bundle to draw on insights of resources.Item The Role of Protein Change (Cellular Protein Loss and Denaturation) in Determining Outcomes of Heating, Cryotherapy and Irreversible Electroporation(2018-04) Liu, FengAtrial fibrillation currently affects millions of people in the US alone. Focal therapy is an increasingly attractive treatment for atrial fibrillation that avoids the debilitating effects of drugs for disease control. Perhaps the most widely used focal therapy for atrial fibrillation (AF) is heat-based radiofrequency (heating), although cryotherapy (cryo) is rapidly replacing it due to a reduction in side effects and positive clinical outcomes. A third focal therapy, irreversible electroporation (IRE), is also being considered in some settings. This study was designed to help guide treatment thresholds and compare mechanism of action across heating, cryo, and IRE. Testing was undertaken on HL-1 cells, a well-established cardiomyocyte cell line, to assess injury thresholds for each treatment method. Cell viability, as assessed by Hoechst and PI staining, was found to be minimal after exposure to temperatures ≤-40 °C (cryo), ≥60 °C (heating), and when field strengths ≥1500 V/cm (IRE) were used. Viability was then correlated to protein denaturation fraction (PDF) as assessed by Fourier Transform Infrared (FTIR) spectroscopy, and protein loss fraction (PLF) as assessed by Bicinchoninic Acid (BCA) assay after the three treatments. These protein changes were assessed both in the supernatant and the pellet of cell suspensions post treatment. We found that dramatic viability loss (≥50%) correlated strongly with ≥12% protein change (PLF, PDF or a combination of the two) in every focal treatment. These studies help in defining both cellular thresholds and protein-based mechanisms of action that can be used to improve focal therapy application for atrial fibrillation.