A5100366053- Astrophysics and Cosmic Phenomena
- Neutrino Physics Research
- Dark Matter and Cosmic Phenomena
- Particle physics theoretical and experimental studies
- Radio Astronomy Observations and Technology
- Advanced Data Storage Technologies
- High-Energy Particle Collisions Research
- Particle Detector Development and Performance
- Gamma-ray bursts and supernovae
- Distributed systems and fault tolerance
- Pulsars and Gravitational Waves Research
- Quantum Chromodynamics and Particle Interactions
- Cloud Computing and Resource Management
- Distributed and Parallel Computing Systems
- Caching and Content Delivery
- Cloud Data Security Solutions
- Radiation Detection and Scintillator Technologies
- Atmospheric Ozone and Climate
- Computational Physics and Python Applications
- Radio Wave Propagation Studies
- Cosmology and Gravitation Theories
- Solar and Space Plasma Dynamics
- Radio, Podcasts, and Digital Media
- Astronomical Observations and Instrumentation
- Peer-to-Peer Network Technologies
University of Wisconsin–Madison
2011-2025
CSSC Offshore & Marine Engineering Company (China)
2024
Southern University and Agricultural and Mechanical College
2018-2024
Clark Atlanta University
2018-2024
University of Geneva
2018-2024
The University of Texas at Arlington
2018-2024
University of Alaska Anchorage
2018-2024
Providence College
2018-2024
Lawrence Berkeley National Laboratory
2024
Istituto Nazionale di Fisica Nucleare, Sezione di Lecce
2023
We present WiscKey, a persistent LSM-tree-based key-value store with performance-oriented data layout that separates keys from values to minimize I/O amplification. The design of WiscKey is highly SSD optimized, leveraging both the sequential and random performance characteristics device. demonstrate advantages microbenchmarks YCSB workloads. Microbenchmark results show 2.5 × 111 faster than LevelDB for loading database (with significantly better tail latencies) 1.6 14 lookups. RocksDB in all six
We conduct a comprehensive study of file-system code evolution. By analyzing eight years Linux changes across 5079 patches, we derive numerous new (and sometimes surprising) insights into the development process; our results should be useful for both file systems themselves as well improvement bug-finding tools.
We conduct a comprehensive study of file-system code evolution. By analyzing eight years Linux changes across 5079 patches, we derive numerous new (and sometimes surprising) insights into the development process; our results should be useful for both file systems themselves as well improvement bug-finding tools.
The A(Q^2) structure function in elastic electron-deuteron scattering was measured at six momentum transfers Q^2 between 0.66 and 1.80 (GeV/c)^2 Hall C Jefferson Laboratory. scattered electrons recoil deuterons were detected coincidence, a fixed deuteron angle of 60.5 degrees. These new precise measurements resolve discrepancies older sets data. They put significant constraints on existing models the electromagnetic structure, strength isoscalar meson exchange currents.
In sensor networks we aim to achieve global objectives through local decisions at each node, based only on data available in the node's neighborhood. this paper, diffuse information away from source nodes holding desired data, so as establish potentials that allow network queries navigate towards and reach these sources greedy decisions, following gradients. We compute by solving for a discrete approximation partial differential equation over appropriate neighborhoods, simple iteration can...
We introduce IceFS, a novel file system that separates physical structures of the system. A new abstraction, cube, is provided to enable grouping files and directories inside physically isolated container. show three major benefits cubes within IceFS: localized reaction faults, fast recovery, concurrent file-system updates. demonstrate these VMware-based virtualized environment Hadoop distributed Results our prototype can significantly improve availability performance, sometimes by an order...
Recent research has shown that applications often incorrectly implement crash consistency. We present the Crash-Consistent File System (ccfs), a file system improves correctness of application-level consistency protocols while maintaining high performance. A key idea in ccfs is abstraction stream . Within stream, updates are committed program order, improving correctness; across streams, there no ordering restrictions, enabling scheduling flexibility and empirically demonstrate running atop...
Virtualization provides the possibility of whole machine migration and thus enables a new form fault tolerance that is completely transparent to applications operating systems. While initial prototypes show promise, virtualization-based fault-tolerant architecture still experiences substantial performance overhead especially for data-intensive workloads. The main challenge how synchronize memory states Master Slave in way minimizes end-to-end impact on application performance. This paper...
There has been increasing interest in deploying data reduction techniques primary storage systems. This paper analyzes large datasets four typical enterprise environments to find patterns that can suggest good design choices for such The overall opportunity is evaluated deduplication and compression, separately combined, then in-depth analysis presented focusing on frequency, clustering other the collected data. results ways enhance performance reduce resource requirements system cost while...
Mariner is an iSCSI-based storage system that designed to provide comprehensive data protection on commodity ATA disk and gigabit Ethernet technologies while offering the same performance as those without any such protection. In particular, supports continuous (CDP) allows every update within a time window be undoable, local/remote mirroring guard against machine/site failures. To minimize overhead associated with CDP, employs modified track-based logging technique unifies long-term required...
The growing popularity of hosted storage services and shared infrastructure in data centers is driving the recent interest resource management QoS systems. bursty nature workloads raises significant performance provisioning challenges, leading to increased infrastructure, management, energy costs. We present a novel dynamic workload shaping framework handle workloads, where arrival stream dynamically decomposed isolate its bursts, then rescheduled exploit available slack. show how...
In this White Paper, we present recommendations for the scientific community and funding agencies to foster infrastructure a collaborative multi-messenger time-domain astronomy (MMA/TDA) ecosystem. MMA/TDA is poised breakthrough discoveries in coming decade. much same way that expanding beyond optical bandpass revealed entirely new unexpected discoveries, cosmic messengers light (i.e., gravitational waves, neutrinos, rays) open windows answer some of most fundamental questions...
Topology optimization is extensively employed in engineering design. However, applying deep learning to solve topology problems faces challenges owing limited training data and model adaptability various boundary conditions. To address these challenges, this research has a solid isotropic material with penalization method produce dataset of 400,000 2D samples covering four Each condition includes two resolution datasets. Moreover, an improved DoubleU-Net proposed for real-time optimization,...
Energy consumption is becoming an increasingly important issue in storage systems, especially for high performance data centers and network servers. In this paper, we introduce a family of energy-efficient disk layouts that generalize the mirroring conventional RAID 1 system. The scheme called DiskGroup distributes workload between primary disks secondary based on characteristics workload. We develop analytic model to explore design space compute estimated energy savings as function...
Block-level continuous data protection (CDP) logs every disk block update so that updates within a time window are undoable. Standard file servers and DBMS can enjoy the service offered by block-level CDP without any modification. Unfortunately, no existing systems provide users versioning view on top of versions they maintain. As result, maintain cannot be used as an extension to on-line system with which routinely interact. This paper describes name-based user-level called UVFS is designed...
Mariner is an iSCSI-based storage system that designed to provide comprehensive data protection on commodity ATA disk and gigabit Ethernet technologies while offering the same performance as those without any such protection. In particular, supports continuous (CDP) allows every update within a time window be undoable, local/remote mirroring guard against machine/site failures. To minimize overhead associated with CDP, employs modified track-based logging technique unifies long-term required...