Linbo Long

ORCID: 0000-0003-1966-0714
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Advanced Data Storage Technologies
  • Parallel Computing and Optimization Techniques
  • Caching and Content Delivery
  • Advanced Memory and Neural Computing
  • Cloud Computing and Resource Management
  • Distributed and Parallel Computing Systems
  • Peer-to-Peer Network Technologies
  • Green IT and Sustainability
  • IoT and Edge/Fog Computing
  • Interconnection Networks and Systems
  • Algorithms and Data Compression
  • Data Mining Algorithms and Applications
  • Fuzzy Logic and Control Systems
  • Numerical Methods and Algorithms
  • Cellular Automata and Applications
  • Energy Harvesting in Wireless Networks
  • Digital and Cyber Forensics
  • Customer churn and segmentation
  • Advanced Malware Detection Techniques
  • Complex Network Analysis Techniques
  • Quantum Computing Algorithms and Architecture
  • Ferroelectric and Negative Capacitance Devices
  • Opinion Dynamics and Social Influence
  • Customer Service Quality and Loyalty
  • Digital Rights Management and Security

Chongqing University of Posts and Telecommunications
2016-2025

ORCID
2018

Chongqing University
2013-2016

Ministry of Education of the People's Republic of China
2014-2016

Solid State Drives (SSDs) based on the NVMe Zoned Namespaces (ZNS) interface can notably reduce costs of address mapping, garbage collection, and over-provisioning by dividing storage space into multiple zones for sequential writes random reads. The Log-Structured Merge (LSM) tree, which is extensively used in key-value systems, converts to writes, hence a suitable scenario utilize ZNS SSDs. However, LSM tree associated data significantly varies lifetime due levels merging mechanisms tree....

10.1145/3721287 article EN ACM Transactions on Architecture and Code Optimization 2025-03-04

Smartphones are getting increasingly high-performance with advances in mobile processors and larger main memories to support feature-rich applications. However, the storage subsystem has always been a prohibitive factor that slows down pace of reaching even higher performance while maintaining good user experience. Despite today's smartphones equipped larger-than-ever memories, they consume more energy still run out memory. But slow NAND flash based vetoes possibility swapping---an important...

10.1145/2656045.2656049 article EN 2014-10-12

ZNS SSDs divide the storage space into sequential-write zones, reducing costs of DRAM utilization, garbage collection, and over-provisioning. The feature zones is well-suited for LSM-based databases, where random writes are organized sequential to improve performance. However, current compaction mechanism LSM-tree results in widely varying access frequencies (i.e., hotness) data thus incurs an extreme imbalance distribution erasure counts across zones. significantly limits lifetime SSDs....

10.1145/3637488 article EN ACM Transactions on Architecture and Code Optimization 2023-12-13

Modern NVMe SSDs have been widely deployed in multi-tenant cloud computing environments or multi-programming systems. When multiple applications concurrently access one SSD hardware, unfairness within the shared will slow down application significantly and lead to a violation of service level objectives. However, traditional data cache management mainly focuses on improving hit ratio, which causes contention sacrifices fairness among applications. In this paper, we propose DRAM-based...

10.1109/tpds.2022.3143295 article EN IEEE Transactions on Parallel and Distributed Systems 2022-01-14

The NVMe Zoned Namespace (ZNS) is a new type of storage interface, which divides logical address space into fixed-size zones, and each zone strictly follows sequential write constraint with pointer. Owing to the ZNS, I/O requests would not be scheduled arbitrarily like traditional SSDs block interface. When multiple applications concurrently access one ZNS SSD hardware, deteriorates blocking causes huge unfairness. To resolve problem, we propose self-balance scheduling dedicated for SSDs,...

10.1109/tcad.2022.3232997 article EN IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 2022-12-30

10.1109/asp-dac58780.2024.10473944 article EN 2022 27th Asia and South Pacific Design Automation Conference (ASP-DAC) 2024-01-22

The Zoned Namespace (ZNS) SSD is a newly introduced storage device and provides several new ZNS commands to upper-level applications. Zone-reset command one of the erase all flash blocks within zone. Since data grouped erased in zone units, SSDs are widely used LSM-tree-based stores. However, basic invalidated unit LSM-tree an SST/WAL file, which mismatches erasing SSD. Placing different files same zone, on faces dramatic space amplification extensive migration problems.

10.1145/3673038.3673096 article EN 2024-08-08

Smartphones are becoming increasingly energy-hungry to support feature-rich applications, posing a lot of pressure on battery lifetime and making energy consumption non-negligible issue. In particular, dynamic random access memory (DRAM)-based main subsystem is major contributor the mobile devices. this paper, we propose direct read (DR). Swap, an energy-efficient in-memory paging design reduce in smartphones. DR. adopt emerging nonvolatile (NVM) use it as swap area. Utilizing NVMs...

10.1109/tcad.2015.2512904 article EN IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 2015-12-29

To achieve efficient file access to the system backend, clients employ various intricate optimization techniques, such as local data/metadata caching and direct data access. However, these techniques impose a significant load on host CPU, posing substantial challenges valuable CPU resources.

10.1145/3673038.3673123 article EN cc-by 2024-08-08

Virtualization offers significant benefits, such as better isolation and security for mobile systems. However, the limited amount of memory virtualization's memory-demanding nature make it challenging to virtualize systems efficiently. In this paper, we utilize morphable resistive memories design a high-performance system with an extensible space. With memories, simple effective page management technique, Balloonfish, is proposed convert cell state between multilevel single-level achieving...

10.1109/tcad.2015.2495264 article EN IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 2015-10-27

Virtualization offers significant benefits such as better isolation and security for mobile systems. However, the limited amount of memory virtualization's memory-demanding nature makes it challenging to virtualize systems efficiently. In this paper, we utilize morphable resistive memories design a high-performance system with extensible space. With memory, convert cell state between multi-level single-level achieve balance performance Our evaluation based on Samsung Exynos 5250 SoC real...

10.1109/aspdac.2015.7059025 article EN 2015-01-01

10.1016/j.physa.2018.05.149 article EN Physica A Statistical Mechanics and its Applications 2018-06-21

Phase change memory (PCM) has emerged as a promising candidate to replace DRAM in embedded systems. However, it can only sustain limited number of write operations. To solve this issue, paper proposes novel and effective wear-leveling technique software level prolong the lifetime PCM-based A polynomial-time algorithm, Multi-Space Wear Leveling Algorithm (MWL), is proposed achieve wear-leveling. The experimental results show our greatly extend systems compared with previous work. Compared...

10.1109/rtcsa.2013.6732213 article EN 2013-08-01

Similar to traditional NAND flash memory, triple-level cell (TLC) memory is used as secondary storage meet the fast growing demands on capacity. TLC exhibits attractive features such shock resistance, high density, low cost, non-volatility and access latency natures. However, also has some extra limitations, write disturbance, performances very limited cycles compared single-level (SLC) memory. In this paper, we propose a workload-aware translation layer, named TLC-FTL, for TLC/SLC dual-mode...

10.1109/hpcc-css-icess.2015.263 article EN 2015-08-01

Scratchpad memory (SPM) is widely utilized in many embedded systems as a software- controlled on-chip to replace the traditional cache. New non-volatile (NVM) has emerged promising candidate SRAM SPM, due its significant benefits, such low-power consumption and high performance. In particular, several representative NVMs, PCM, ReRAM, STT-RAM can build multiple-level cells (MLC) achieve even higher density. Nevertheless, this triggers off energy overhead longer access latency compared with...

10.1109/access.2018.2872762 article EN cc-by-nc-nd IEEE Access 2018-01-01

Mobile applications are becoming more and powerful but also dependent on large main memories, which consume a portion of system energy. Swapping to byte-addressable, non-volatile memory (NVRAM) is promising solution this problem. However, most NVRAMs have limited write endurance. To make it practical, the design an NVRAM based swapping must consider In paper, we target at prolonging lifetime swap area in mobile devices. Different from traditional wisdom, such as wear leveling hot/cold data...

10.5555/2755753.2757117 article EN Design, Automation, and Test in Europe 2015-03-09

High bit-density flash memories, such as triple-level cell (TLC) and quad-level (QLC), have been widely used in memory-based storage systems, offering significantly high capacity. However, these memories suffer from asymmetric access performance on the different pages that sharing same physical cells. Meanwhile, 3-D memory adopts stacking technology to increase capacity reduce cost per bit. The unit can be reprogrammed many times long voltage increases. reprogramming is also an effective...

10.1109/tcad.2022.3210390 article EN IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 2022-09-28

ZNS SSDs shift the responsibility of garbage collection (GC) to host. However, data migration in GC needs move host's buffer first and write back new location, resulting an unnecessary end-to-end transfer overhead. Moreover, due pre-configured mapping between zones blocks, perform a large number block-to-block migrations zones. To address these issues, this paper proposes simple efficient method, called IS-AR, with in-storage remapping. Based on full-stack SSD emulator, our evaluation shows...

10.23919/date56975.2023.10137231 article EN Design, Automation & Test in Europe Conference & Exhibition (DATE), 2015 2023-04-01

Abstract This paper addresses the feature selection problem for gene expression profiles. Feature based on fuzzy neighborhood rough sets is very important in The process of extracting key features profiles may have two disadvantages: (1) it generate many redundant features, thus reducing classification accuracy, and (2) ignore some information. To address above problems, this proposes a joint entropy model selection; adopts nonnegative principle to evaluate importance candidate gene. Based...

10.1007/s40747-023-01138-9 article EN cc-by Complex & Intelligent Systems 2023-07-17
Coming Soon ...