Towards the Design of Efficient and Consistent Index Structure with Minimal Write Activities for Non-Volatile Memory

Sha, Edwin Hsing-Mean; Jiang, Weiwen; Dong, Hailiang; Ma, Zhulin; Zhang, Runyu; Chen, Xianzhang; Zhuge, Qingfeng

doi:10.1109/tc.2017.2754381

Cited by 6 publications

(2 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides these, another work [28] refines the sorted linked list by providing a dual contiguous address space, so that it proposes an efficient binary search strategy which improves the performance of query operations. [29] Custom delays/cross-NUMA delays Custom delays/cross-NUMA delays No emulation/cross-NUMA bandwidth…”

Section: Nvm and Persistent Data Structuresmentioning

confidence: 99%

Revisiting Persistent Indexing Structures on Intel Optane DC Persistent Memory

Dong

et al. 2021

J. Comput. Sci. Technol.

View full text Add to dashboard Cite

Persistent indexing structures are proposed in response to emerging non-volatile memory (NVM) to provide high performance yet durable indexes. However, due to the lack of real NVM hardware, many prior persistent indexing structures were evaluated via emulation, which varies a lot across different setups and differs from the real deployment. Recently, Intel has released its Optane DC Persistent Memory Module (PMM), which is the first production-ready NVM. In this paper, we revisit popular persistent indexing structures on PMM and conduct comprehensive evaluations to study the performance differences among persistent indexing structures, including persistent hash tables and persistent trees. According to the evaluation results, we find that Cacheline-Conscious Extendible Hashing (CCEH) achieves the best performance among all evaluated persistent hash tables, and Failure-Atomic ShifT B+-Tree (FAST) and Write Optimal Radix Tree (WORT) perform better than other trees. Besides, we find that the insertion performance of hash tables is heavily influenced by data locality, while the insertion latency of trees is dominated by the flush instructions. We also uncover that no existing emulation methods accurately simulate PMM for all the studied data structures. Finally, we provide three suggestions on how to fully utilize PMM for better performance, including using clflushopt/clwb with sfence instead of clflush, flushing continuous data in a batch, and avoiding data access immediately after it is flushed to PMM.

show abstract

Section: Nvm and Persistent Data Structuresmentioning

confidence: 99%

Revisiting Persistent Indexing Structures on Intel Optane DC Persistent Memory

Dong

et al. 2021

J. Comput. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…The for loop was used to open 2000 simulated vehicle terminals [33,34], and the new vehicle terminal initiated a connection request to the server and sent data. The transmission period, T, of each on-board terminal was 500 ms, and each on-board terminal was sent 1000 times.…”

Section: Platform Stress Testmentioning

confidence: 99%

Design and Implementation of the Remote Operation and Maintenance Platform for the Combine Harvester

et al. 2022

View full text Add to dashboard Cite

To meet the needs of remotely operating and maintaining combine harvesters in view of the problems of extensive operation, inefficient operation and maintenance, and backward management of combine harvesters, a remote operation and maintenance platform for the combine harvester has been designed. The platform mainly includes modules, such as data monitoring, fault prediction, fault diagnosis, and comprehensive operation maintenance. Among them, the data-monitoring module obtains the monitoring data of the vehicle terminal. It actively transmits to the remote operation and maintenance platform to remotely monitor the operation data or status of the combine harvester system. The fault prediction module builds a fault prediction model based on the fault features and data provided by all previous stages. It predicts and evaluates the actual operation of each component of the combine harvester in the foreseeable future, Based on the fault diagnosis model, the fault diagnosis module evaluates and classifies the current abnormal state of the combine harvester. The comprehensive operation and maintenance module formulates the corresponding maintenance and repair strategies according to the output results of the fault diagnosis and prediction. Through functional field testing of the platform, the results showed that the platform could perform real-time monitoring, fault prediction, fault diagnosis, and comprehensive operation maintenance for the operation information of the combine harvester. It helped to improve the intelligent operation level, management quality and operation, and maintenance efficiency of the combine harvester.

show abstract