Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 15 of 15 for “"Straggler"”.
-
Mitigating Spark straggler tasks for iterative applications by data re-partitioning
… one problem that affects the running time is the straggler, where a disproportionate long-running task slows down the entire cluster. In this work we present a straggler mitigation technique tailored for applications that run small tasks for many iterations over a large dataset, and implemented …
-
OS INSPIRED COMPLETE KERNEL FUSION
… bandwidth and is especially vulnerable to straggler-induced performance degradation. Focusing on dynamic workloads such as Mixture-of-Experts (MoE), we highlight two critical bottlenecks. First, we demonstrate how CPU-driven execution limits the exploitation of task locality and introduces …
-
A panchromatic study of stellar populations in Galactic Globular Clusters
… to investigate (i) how common the bimodal Blue Straggler Stars (BSS) radial distribution is in stellar clusters and (ii) which are the physical processes that can produce this bimodality. We discuss possible relations between the properties of the BSS radial distribution and the dynamical state …
-
Distributed Machine Learning in Heterogeneous Edge Networks
… learning poses unique challenges, including the straggler problem (i.e., where slow devices hinder training progress), communication overhead (i.e., the high cost of transferring data between devices), and device heterogeneity (i.e., variability in device capabilities). This dissertation explores …
-
Speeding Up Coded Distributed Machine Learning
… is challenging in practice due to the issue of stragglers. Stragglers are the nodes that temporarily underperform for multiple reasons, such as resource contention or network error, significantly delaying the training progress. Assigning identical tasks to multiple workers is a straightforward …
-
New techniques to lower the tail latency in stream processing systems
… timeout coupled with selective replay to catch straggler tuples; (2) shared queues among different tasks of the same operator to reduce overall queueing delay; (3) latency feedback-based load balancing, intended to mitigate heterogenous scenarios. We have implemented these techniques in Apache …
-
Incentive mechanism design in blockchain-based federated learning over edge clouds
… First, this research addresses the latency and straggler issues inherent in synchronous FL by proposing an incentive mechanism for a semi-asynchronous blockchain-based federated edge learning system. The resource pricing interaction between edge servers and task publishers is modeled as a …
-
Reinforcement Learning in Time-Varying Systems: an Empirical Study
… We apply our framework to two systems problems: straggler mitigation and adaptive video streaming, and evaluate it against a variety of alternative approaches using real-world and synthetic data. We show that each component of our framework is necessary to cope with non-stationarity.
-
Algebraic approaches for coded caching and distributed computing
… caching for large scale content distribution and straggler mitigation within distributed computation.</p> <p>Caching is a popular technique for facilitating large scale content delivery over the Internet. Traditionally, caching operates by storing popular content closer to the end users. Recent …
-
A Dynamic Resource Management Framework and Reconfiguration Strategies for Cloud-native Bulk Synchronous Parallel Applications
… imbalance across supersteps can trigger the straggler effect, where faster tasks wait at synchronization barriers for slower ones, increasing overall execution time. Existing BSP resource management approaches typically assume static workloads and reuse a single configuration throughout …
-
Galactic Dark Matter
… distributions of blue horizontal-branch and blue straggler stars. It is found that this analysis allows for a wide range of parameters for the dark matter. The implications for direct and indirect detection of dark matter are discussed in detail.</p><p>In the appendices, two additional projects …
-
Synchronization-avoiding graph algorithms and runtime aspects
… can suffer performance bottlenecks due to the straggler effect induced by global and vertex-centric barriers. In the effort to eliminate barriers, we design and study unordered, data-driven graph algorithms, relying on optimistic (speculative) execution. Our design of algorithms allows work to …
-
A spectroscopic and photometric study of MSP companions in Galactic Globular Clusters
… (MSP), low mass X-ray binaries(LMXB) or Blue Straggler Stars (BSS). In particular MSPs are formed in binary systems containing a Neutron Star which is spun up through mass accretion from the evolving companion (e.g. Bhattacharia & van den Heuvel 1991). The final stage of this recycling process …
-
Scalability Analysis and Optimization for Large-Scale Deep Learning
… uncoordinated manner, which results in straggler processes and consequently computational imbalance, further limiting scalability. This thesis studies a subset of the large body of data movement and data processing challenges that exist in modern DL frameworks. For the first study, we …
-
Efficient Network Systems Design for Machine Learning
… directly inside the network, and in-network straggler mitigation, which utilizes Trio timer threads to detect and address stragglers. We prototype Trio-ML on a testbed with three real DNN models (ResNet50, DenseNet161, and VGG11) to demonstrate its effectiveness in mitigating stragglers while …