A5100345233- Advanced Data Storage Technologies
- Caching and Content Delivery
- Cell Image Analysis Techniques
- Advanced Fluorescence Microscopy Techniques
- Distributed systems and fault tolerance
- Cloud Data Security Solutions
- Single-cell and spatial transcriptomics
- Advanced Biosensing Techniques and Applications
- Photosynthetic Processes and Mechanisms
- Advanced Electron Microscopy Techniques and Applications
- Error Correcting Code Techniques
- Parallel Computing and Optimization Techniques
Huazhong University of Science and Technology
2014-2022
Wuhan National Laboratory for Optoelectronics
2014-2022
In backup systems, the chunks of each are physically scattered after deduplication, which causes a challenging fragmentation problem. We observe that comes into sparse and out-of-order containers. The container decreases restore performance garbage collection efficiency, while if cache is small. order to reduce fragmentation, we propose History-Aware Rewriting algorithm (HAR) Cache-Aware Filter (CAF). HAR exploits historical information in systems accurately identify containers, CAF...
Light-sheet fluorescence microscopy (LSFM) allows volumetric live imaging at high-speed and with low photo-toxicity. Various LSFM modalities are commercially available, but their size cost limit access by the research community. A new method, termed sub-voxel-resolving (SVR) light-sheet add-on (SLAM), is presented to enable fast, resolution-enhanced from a conventional wide-field microscope. This method contains two components: miniature device regular microscopes, which horizontal laser...
Erasure codes are widely used in distributed storage systems to prevent data loss. Traditional erasure suffer from a typical repair-bandwidth problem which the amount of required reconstruct lost data, referred as repair bandwidth, is often far more than theoretical minimum. While many novel have been proposed recent years reduce these either require extra capacity and computation overhead or only applicable some special cases. To address weaknesses existing solutions problem, we propose Z...
Erasure codes are widely used in distributed storage systems to prevent data loss. Traditional suffer from a typical repair-bandwidth problem which the amount of required reconstruct lost data, referred as repair bandwidth, is often far more than theoretical minimum. While many novel have been proposed recent years reduce these either require extra and computation overhead or only applicable some special cases. To address weaknesses existing solutions problem, we propose Z Codes, general...
Erasure codes are widely used in practical storage systems to prevent disk failure and data loss. However, these require excessive I/Os network traffic for recovering unavailable data. As a result, the recovery performance of erasure is suboptimal. Among all codes, Minimum Storage Regenerating (MSR) can achieve optimal repair bandwidth under minimum during recovery, but some open issues remain be addressed before applying them real systems. In this paper, we present Hybrid Codes (Hybrid-RC),...
Erasure codes are widely used in modern distributed storage systems to prevent data loss and server failures. Regenerating a class of erasure that trades efficiency computation for repair bandwidth reduction. However, their non-unified coding parameters huge overhead prohibit applications. Hence, we first propose family Functional Codes (FRCs) with uncoded repair, balancing general parameters. FRCs take advantage heuristic algorithm, which makes efforts employ as little possible. Second,...
Light-sheet fluorescence microscopy (LSFM) allows volumetric live imaging at high-speed and with low photo-toxicity. Various LSFM modalities are commercially available, but their size cost limit access by the research community. A new method, termed sub-voxel-resolving light-sheet add-on (SLAM), is presented to enable fast, resolution-enhanced from a conventional wide-field microscope. This method contains two components: miniature device regular microscopes, which horizontal laser...
Abstract Long-term visualization of the dynamic organelle-organelle or protein-organelle interactions throughout three-dimensional space whole live cells is essential to better understand their functions, but this task remains challenging due limitations existing fluorescence microscopy techniques, such as an insufficient axial resolution, low volumetric imaging rate, and photobleaching. Here, we present combination a progressive deep-learning superresolution strategy with...