The meta-heuristic algorithm is a global probabilistic search algorithm for the iterative solution.It has good performance in global optimization fields such as maximization.In this paper,a new adaptive parameter stra...The meta-heuristic algorithm is a global probabilistic search algorithm for the iterative solution.It has good performance in global optimization fields such as maximization.In this paper,a new adaptive parameter strategy and a parallel communication strategy are proposed to further improve the Cuckoo Search(CS)algorithm.This strategy greatly improves the convergence speed and accuracy of the algorithm and strengthens the algorithm’s ability to jump out of the local optimal.This paper compares the optimization performance of Parallel Adaptive Cuckoo Search(PACS)with CS,Parallel Cuckoo Search(PCS),Particle Swarm Optimization(PSO),Sine Cosine Algorithm(SCA),Grey Wolf Optimizer(GWO),Whale Optimization Algorithm(WOA),Differential Evolution(DE)and Artificial Bee Colony(ABC)algorithms by using the CEC-2013 test function.The results show that PACS algorithmoutperforms other algorithms in 20 of 28 test functions.Due to the superior performance of PACS algorithm,this paper uses it to solve the problem of the rectangular layout.Experimental results show that this scheme has a significant effect,and the material utilization rate is improved from89.5%to 97.8%after optimization.展开更多
The penetration rate of distributed generation is gradually increasing in the distribution system concerned.This is creating new problems and challenges in the planning and operation of the system.The intermittency an...The penetration rate of distributed generation is gradually increasing in the distribution system concerned.This is creating new problems and challenges in the planning and operation of the system.The intermittency and variability of power outputs from numerous distributed renewable generators could significantly jeopardize the secure operation of the distribution system.Therefore,it is necessary to assess the hosting capability for intermittent distributed generation by a distribution system considering operational constraints.This is the subject of this study.An assessment model considering the uncertainty of generation outputs from distributed generators is presented for this purpose.It involves different types of regulation or control functions using on-load tap-changers(OLTCs),reactive power compensation devices,energy storage systems,and the reactive power support of the distributed generators employed.A robust optimization model is then attained It is solved by Bertsimas robust counterpart through GUROBI solver.Finally,the feasibility and efficiency of the proposed method are demonstrated by a modified IEEE 33-bus distribution system.In addition,the effects of the aforementioned regulation or control functions on the enhancement of the hosting capability for intermittent distributed generation are examined.展开更多
The search for fast radio bursts(FRBs) is a hot topic in current radio astronomy studies. In this work, we carry out a single pulse search with a very long baseline interferometry(VLBI) pulsar observation data set usi...The search for fast radio bursts(FRBs) is a hot topic in current radio astronomy studies. In this work, we carry out a single pulse search with a very long baseline interferometry(VLBI) pulsar observation data set using both auto spectrum and cross spectrum search methods. The cross spectrum method,first proposed in Liu et al., maximizes the signal power by fully utilizing the fringe phase information of the baseline cross spectrum. The auto spectrum search method is based on the popular pulsar software package PRESTO, which extracts single pulses from the auto spectrum of each station. According to our comparison, the cross spectrum method is able to enhance the signal power and therefore extract single pulses from data contaminated by high levels of radio frequency interference(RFI), which makes it possible to carry out a search for FRBs in regular VLBI observations when RFI is present.展开更多
Despite the multifaceted advantages of cloud computing,concerns about data leakage or abuse impedes its adoption for security-sensi tive tasks.Recent investigations have revealed that the risk of unauthorized data acc...Despite the multifaceted advantages of cloud computing,concerns about data leakage or abuse impedes its adoption for security-sensi tive tasks.Recent investigations have revealed that the risk of unauthorized data access is one of the biggest concerns of users of cloud-based services.Transparency and accountability for data managed in the cloud is necessary.Specifically,when using a cloudhost service,a user typically has to trust both the cloud service provider and cloud infrastructure provider to properly handling private data.This is a multi-party system.Three particular trust models can be used according to the credibility of these providers.This pa per describes techniques for preventing data leakage that can be used with these different models.展开更多
Given large-scale modern power systems with pow er electronic converters,the numerical simulation with subsyn chronous oscillation(SSO)faces great challenges in engineering practice due to sharply enlarged modeling sc...Given large-scale modern power systems with pow er electronic converters,the numerical simulation with subsyn chronous oscillation(SSO)faces great challenges in engineering practice due to sharply enlarged modeling scale and high com putational burden.To reduce the modeling scale,network parti tion and equivalent becomes a vital technique in numerical sim ulations.Although several methods have been developed for net work equivalent,a generally accepted rule for network parti tion is still required.This paper proposes that the system can be partitioned into three parts,i.e.,the internal,the middle,and the external subsystems,in which the internal subsystem consists of all power electronic components,the middle subsys tem includes those selected AC dynamic components with de tailed models,and the remaining components and buses consti tute the external subsystem.The external subsystem is further represented by an equivalent RLC network determined by the frequency dependent network equivalent(FDNE)method.In the proposed method,the observability index and the electrical distance index are used to identify the interface between the middle and the external subsystems.Case studies based on a modified Hydro-Quebec system are used to verify the effective ness of the proposed method.展开更多
A storage system is the core of a computer,and plays an important role in the sustainable development of emerging strategic industries,such as artificial intelligence,big data,cloud computing,and the Internet of Thing...A storage system is the core of a computer,and plays an important role in the sustainable development of emerging strategic industries,such as artificial intelligence,big data,cloud computing,and the Internet of Things.Storage stack access is a major factor restricting the performance of data-intensive systems because of the increasing performance of processors and network devices.展开更多
Chang’E-3 landed on the east of Sinus Iridum area on December 14,2013,performing China’s first successful soft landing on the lunar surface.We present the results on precision orbit determination and positioning of ...Chang’E-3 landed on the east of Sinus Iridum area on December 14,2013,performing China’s first successful soft landing on the lunar surface.We present the results on precision orbit determination and positioning of the lander and the rover.We describe the data,modeling,and methods used to achieve position knowledge over the period December 2–21,2014.In addition to the radiometric X-band range and Doppler tracking data,delta differential one-way ranging data are also used in the calculation,which show that they strongly improve the accuracy of the orbit reconstruction.Total position overlap differences are about 20 and 30 m for the 100 km 9 100 km and100 km9 15 km lunar orbit,respectively,increased by*50%with respect to CE-2 and at the same level as other lunar spacecrafts of recent era such as SELENE and lunar reconnaissance orbiter(LRO).The position error of the soft landing trajectory is less than 100 m.A kinematic statistical method is applied to determine the position of the lander and relative position of the rover with respect to the lander.The position difference of the lander is better than50 m compared to LRO photograph result.Compared with the delta very long baseline interferometry(VLBI)group delay between the lander and the rover,the delta VLBI phase delay can improve the relative position of the rover from*1,000 to*1 m.展开更多
Weibo is the Twitter counterpart in China that has attracted hundreds of millions of users. We crawled an almost complete Weibo user network that contains 222 million users and 27 billion links in 2013. This paper ana...Weibo is the Twitter counterpart in China that has attracted hundreds of millions of users. We crawled an almost complete Weibo user network that contains 222 million users and 27 billion links in 2013. This paper analyzes the structural properties of this network, and compares it with a Twitter user network. The topological properties we studied include the degree distributions, connected components, distance distributions, reciprocity,clustering coefficient, Page Rank centrality, and degree assortativity. We find that Weibo users have a higher diversity index, higher Gini index, but a lower reciprocity and clustering coefficient for most of the nodes. A surprising observation is that the reciprocity of Weibo is only about a quarter of the reciprocity of the Twitter user network. We also show that Weibo adoption rate correlates with economic development positively, and Weibo network can be used to quantify the connections between provinces and regions in China. In particular, point-wise mutual information is shown to be accurate in quantifying the strength of connections. We developed an interactive analyzing software framework for this study, and released the data and code online.展开更多
Resistive Random Access Memory(ReRAM)-based neural network accelerators have potential to surpass their digital counterparts in computational efficiency and performance.However,design of these accelerators faces a num...Resistive Random Access Memory(ReRAM)-based neural network accelerators have potential to surpass their digital counterparts in computational efficiency and performance.However,design of these accelerators faces a number of challenges including imperfections of the Re RAM device and a large amount of calculations required to accurately simulate the former.We present XB-SIM,a simulation framework for Re RAM-crossbar-based Convolutional Neural Network(CNN)accelerators.XB-SIM can be flexibly configured to simulate the accelerator’s structure and clock-driven behaviors at the architecture level.This framework also includes an Re RAM-aware Neural Network(NN)training algorithm and a CNN-oriented mapper to train an NN and map it onto the simulated design efficiently.Behavior of the simulator has been verified by the corresponding circuit simulation of a real chip.Furthermore,a batch processing mode of the massive calculations that are required to mimic the behavior of Re RAM-crossbar circuits is proposed to fully apply the computational concurrency of the mapping strategy.On CPU/GPGPU,this batch processing mode can improve the simulation speed by up to 5.02 or 34.29.Within this framework,comprehensive architectural exploration and end-to-end evaluation have been achieved,which provide some insights for systemic optimization.展开更多
Inspired by real biological neural models,Spiking Neural Networks(SNNs)process information with discrete spikes and show great potential for building low-power neural network systems.This paper proposes a hardware imp...Inspired by real biological neural models,Spiking Neural Networks(SNNs)process information with discrete spikes and show great potential for building low-power neural network systems.This paper proposes a hardware implementation of SNN based on Field-Programmable Gate Arrays(FPGA).It features a hybrid updating algorithm,which combines the advantages of existing algorithms to simplify hardware design and improve performance.The proposed design supports up to 16384 neurons and 16.8 million synapses but requires minimal hardware resources and archieves a very low power consumption of 0.477 W.A test platform is built based on the proposed design using a Xilinx FPGA evaluation board,upon which we deploy a classification task on the MNIST dataset.The evaluation results show an accuracy of 97.06%and a frame rate of 161 frames per second.展开更多
Brain-inspired computing refers to computational models,methods,and systems,that are mainly inspired by the processing mode or structure of brain.A recent study proposed the concept of"neuromorphic completeness&q...Brain-inspired computing refers to computational models,methods,and systems,that are mainly inspired by the processing mode or structure of brain.A recent study proposed the concept of"neuromorphic completeness"and the corresponding system hierarchy,which is helpful to determine the capability boundary of brain-inspired computing system and to judge whether hardware and software of brain-inspired computing are compatible with each other.As a position paper,this article analyzes the existing brain-inspired chips design characteristics and the current so-called"general purpose"application development frameworks for brain-inspired computing,as well as introduces the background and the potential of this proposal.Further,some key features of this concept are presented through the comparison with the Turing completeness and approximate computation,and the analyses of the relationship with"general-purpose"brain-inspired computing systems(it means that computing systems can support all computable applications).In the end,a promising technical approach to realize such computing systems is introduced,as well as the on-going research and the work foundation.We believe that this work is conducive to the design of extensible neuromorphic complete hardware-primitives and the corresponding chips.On this basis,it is expected to gradually realize"general purpose"brain-inspired computing system,in order to take into account the functionality completeness and application efficiency.展开更多
Multi-Clock Snapshot Isolation(MCSI)is a concurrency control mechanism that implements snapshot isolation on a single-layer Non-Volatile Memory(NVM)database.It stores a single copy of data by using multi-version stora...Multi-Clock Snapshot Isolation(MCSI)is a concurrency control mechanism that implements snapshot isolation on a single-layer Non-Volatile Memory(NVM)database.It stores a single copy of data by using multi-version storage to ensure durability and runtime access.With multi-clock transaction timestamp assignment,MCSI can efficiently generate snapshots with vector clocks and use per-thread transaction status arrays to identify uncommitted versions in NVM.For evaluation,we compared MCSI with the PostgreSQL-style concurrency control used in the single-layer NVM database N2DB.The maximum transaction throughput of MCSI is 101%–195%higher than that of N2DB for the YCSB workloads,and 25%–49%higher for the TPC-C workloads.Moreover,the transaction latency of MCSI remains relatively stable as the thread count increases.With 18 worker threads,the average transaction latency of MCSI is 65%–84%lower than that of N2DB for the YCSB workloads and 16%–43%lower for the TPC-C workloads.展开更多
Non-Volatile Memory(NVM) offers byte-addressability and persistency. Because NVM can be plugged into memory and provide low latency, it offers a new opportunity to build new database systems with a single-layer storag...Non-Volatile Memory(NVM) offers byte-addressability and persistency. Because NVM can be plugged into memory and provide low latency, it offers a new opportunity to build new database systems with a single-layer storage design. A single-layer NVM-Native DataBase(N2 DB) provides zero copy and log freedom. Hence, all data are stored in NVM and there is no extra data duplication and logging during execution. N2 DB avoids complex data synchronization and logging overhead in the two-layer storage design of disk-oriented databases and in-memory databases. Garbage Collection(GC) is critical in such an NVM-based database because memory leaks on NVM are durable. Moreover, data recovery is equally essential to guarantee atomicity, consistency, isolation, and durability properties. Without logging, it is a great challenge for N2 DB to restore data to a consistent state after crashes and recoveries. This paper presents the GC and data recovery mechanisms for N2 DB. Evaluations show that the overall performance of N2 DB is up to 3:6 higher than that of InnoDB. Enabling GC reduces performance by up to 10%,but saves storage space by up to 67%. Moreover, our data recovery requires only 0:2% of the time and half of the storage space of InnoDB.展开更多
基金funded by the NationalKey Research and Development Program of China under Grant No.11974373.
文摘The meta-heuristic algorithm is a global probabilistic search algorithm for the iterative solution.It has good performance in global optimization fields such as maximization.In this paper,a new adaptive parameter strategy and a parallel communication strategy are proposed to further improve the Cuckoo Search(CS)algorithm.This strategy greatly improves the convergence speed and accuracy of the algorithm and strengthens the algorithm’s ability to jump out of the local optimal.This paper compares the optimization performance of Parallel Adaptive Cuckoo Search(PACS)with CS,Parallel Cuckoo Search(PCS),Particle Swarm Optimization(PSO),Sine Cosine Algorithm(SCA),Grey Wolf Optimizer(GWO),Whale Optimization Algorithm(WOA),Differential Evolution(DE)and Artificial Bee Colony(ABC)algorithms by using the CEC-2013 test function.The results show that PACS algorithmoutperforms other algorithms in 20 of 28 test functions.Due to the superior performance of PACS algorithm,this paper uses it to solve the problem of the rectangular layout.Experimental results show that this scheme has a significant effect,and the material utilization rate is improved from89.5%to 97.8%after optimization.
基金the Scientific and Technological Project of SGCC Headquarters entitled“Smart Distribution Network and Ubiquitous Power Internet of Things Integrated Development Collaborative Planning Technology Research”(5400-201956447A-0-0-00).
文摘The penetration rate of distributed generation is gradually increasing in the distribution system concerned.This is creating new problems and challenges in the planning and operation of the system.The intermittency and variability of power outputs from numerous distributed renewable generators could significantly jeopardize the secure operation of the distribution system.Therefore,it is necessary to assess the hosting capability for intermittent distributed generation by a distribution system considering operational constraints.This is the subject of this study.An assessment model considering the uncertainty of generation outputs from distributed generators is presented for this purpose.It involves different types of regulation or control functions using on-load tap-changers(OLTCs),reactive power compensation devices,energy storage systems,and the reactive power support of the distributed generators employed.A robust optimization model is then attained It is solved by Bertsimas robust counterpart through GUROBI solver.Finally,the feasibility and efficiency of the proposed method are demonstrated by a modified IEEE 33-bus distribution system.In addition,the effects of the aforementioned regulation or control functions on the enhancement of the hosting capability for intermittent distributed generation are examined.
基金funded by the National Natural Science Foundation of China (Grant Nos. 11373061, 11573057 and U1631122)the Key Laboratory of Radio Astronomy of Chinese Academy of Sciences (CAS)+2 种基金Shanghai Key Laboratory of Space Navigation and Positioning Techniques (ZZXT-201702)the CAS Key Technology Talent Programthe National R&D Infrastructure and Facility Development Program of China, “Fundamental Science Data Sharing Platform” (DKA2017-12-02-09)
文摘The search for fast radio bursts(FRBs) is a hot topic in current radio astronomy studies. In this work, we carry out a single pulse search with a very long baseline interferometry(VLBI) pulsar observation data set using both auto spectrum and cross spectrum search methods. The cross spectrum method,first proposed in Liu et al., maximizes the signal power by fully utilizing the fringe phase information of the baseline cross spectrum. The auto spectrum search method is based on the popular pulsar software package PRESTO, which extracts single pulses from the auto spectrum of each station. According to our comparison, the cross spectrum method is able to enhance the signal power and therefore extract single pulses from data contaminated by high levels of radio frequency interference(RFI), which makes it possible to carry out a search for FRBs in regular VLBI observations when RFI is present.
基金supported by National Basic Research (973) Program of China (2011CB302505)Natural Science Foundation of China (61373145, 61170210)+1 种基金National High-Tech R&D (863) Program of China (2012AA012600,2011AA01A203)Chinese Special Project of Science and Technology (2012ZX01039001)
文摘Despite the multifaceted advantages of cloud computing,concerns about data leakage or abuse impedes its adoption for security-sensi tive tasks.Recent investigations have revealed that the risk of unauthorized data access is one of the biggest concerns of users of cloud-based services.Transparency and accountability for data managed in the cloud is necessary.Specifically,when using a cloudhost service,a user typically has to trust both the cloud service provider and cloud infrastructure provider to properly handling private data.This is a multi-party system.Three particular trust models can be used according to the credibility of these providers.This pa per describes techniques for preventing data leakage that can be used with these different models.
基金supported by the Science and Technology Project of State Grid Zhejiang Electric Power Co.,Ltd.(No.5211JY21N001)the National Natural Science Foundation of China(No.51837004).
文摘Given large-scale modern power systems with pow er electronic converters,the numerical simulation with subsyn chronous oscillation(SSO)faces great challenges in engineering practice due to sharply enlarged modeling scale and high com putational burden.To reduce the modeling scale,network parti tion and equivalent becomes a vital technique in numerical sim ulations.Although several methods have been developed for net work equivalent,a generally accepted rule for network parti tion is still required.This paper proposes that the system can be partitioned into three parts,i.e.,the internal,the middle,and the external subsystems,in which the internal subsystem consists of all power electronic components,the middle subsys tem includes those selected AC dynamic components with de tailed models,and the remaining components and buses consti tute the external subsystem.The external subsystem is further represented by an equivalent RLC network determined by the frequency dependent network equivalent(FDNE)method.In the proposed method,the observability index and the electrical distance index are used to identify the interface between the middle and the external subsystems.Case studies based on a modified Hydro-Quebec system are used to verify the effective ness of the proposed method.
文摘A storage system is the core of a computer,and plays an important role in the sustainable development of emerging strategic industries,such as artificial intelligence,big data,cloud computing,and the Internet of Things.Storage stack access is a major factor restricting the performance of data-intensive systems because of the increasing performance of processors and network devices.
基金supported by the National Natural Science Foundation of China (11073047,11173052)the Science and Technology Commission of Shanghai (12DZ2273300)+2 种基金the National High Technology Research and Development Program of China (2012AA121603)the Planetary Sciences Laboratory of Chinese Academy of Sciencesthe Lunar Exploration Project of China
文摘Chang’E-3 landed on the east of Sinus Iridum area on December 14,2013,performing China’s first successful soft landing on the lunar surface.We present the results on precision orbit determination and positioning of the lander and the rover.We describe the data,modeling,and methods used to achieve position knowledge over the period December 2–21,2014.In addition to the radiometric X-band range and Doppler tracking data,delta differential one-way ranging data are also used in the calculation,which show that they strongly improve the accuracy of the orbit reconstruction.Total position overlap differences are about 20 and 30 m for the 100 km 9 100 km and100 km9 15 km lunar orbit,respectively,increased by*50%with respect to CE-2 and at the same level as other lunar spacecrafts of recent era such as SELENE and lunar reconnaissance orbiter(LRO).The position error of the soft landing trajectory is less than 100 m.A kinematic statistical method is applied to determine the position of the lander and relative position of the rover with respect to the lander.The position difference of the lander is better than50 m compared to LRO photograph result.Compared with the delta very long baseline interferometry(VLBI)group delay between the lander and the rover,the delta VLBI phase delay can improve the relative position of the rover from*1,000 to*1 m.
基金supported by NSERC(Natural Sciences and Engineering Research Council of Canada)Discovery grant(No.RGPIN-2014-04463)the National High-Tech Research and Development(863)Program of China(No.2012AA010903)the National Natural Science Foundation of China(Nos.61433008 and U1435216)
文摘Weibo is the Twitter counterpart in China that has attracted hundreds of millions of users. We crawled an almost complete Weibo user network that contains 222 million users and 27 billion links in 2013. This paper analyzes the structural properties of this network, and compares it with a Twitter user network. The topological properties we studied include the degree distributions, connected components, distance distributions, reciprocity,clustering coefficient, Page Rank centrality, and degree assortativity. We find that Weibo users have a higher diversity index, higher Gini index, but a lower reciprocity and clustering coefficient for most of the nodes. A surprising observation is that the reciprocity of Weibo is only about a quarter of the reciprocity of the Twitter user network. We also show that Weibo adoption rate correlates with economic development positively, and Weibo network can be used to quantify the connections between provinces and regions in China. In particular, point-wise mutual information is shown to be accurate in quantifying the strength of connections. We developed an interactive analyzing software framework for this study, and released the data and code online.
基金supported in part by Beijing Academy of Artificial Intelligence(BAAI)(No.BAAI2019ZD0403)Beijing Innovation Center for Future Chip,Tsinghua Universitythe Science and Technology Innovation Special Zone Project,China。
文摘Resistive Random Access Memory(ReRAM)-based neural network accelerators have potential to surpass their digital counterparts in computational efficiency and performance.However,design of these accelerators faces a number of challenges including imperfections of the Re RAM device and a large amount of calculations required to accurately simulate the former.We present XB-SIM,a simulation framework for Re RAM-crossbar-based Convolutional Neural Network(CNN)accelerators.XB-SIM can be flexibly configured to simulate the accelerator’s structure and clock-driven behaviors at the architecture level.This framework also includes an Re RAM-aware Neural Network(NN)training algorithm and a CNN-oriented mapper to train an NN and map it onto the simulated design efficiently.Behavior of the simulator has been verified by the corresponding circuit simulation of a real chip.Furthermore,a batch processing mode of the massive calculations that are required to mimic the behavior of Re RAM-crossbar circuits is proposed to fully apply the computational concurrency of the mapping strategy.On CPU/GPGPU,this batch processing mode can improve the simulation speed by up to 5.02 or 34.29.Within this framework,comprehensive architectural exploration and end-to-end evaluation have been achieved,which provide some insights for systemic optimization.
基金supported in part by the Beijing Innovation Center for Future Chip,Tsinghua Universityin part by the Science and Technology Innovation Special Zone project,Chinain part by the Tsinghua University Initiative Scientific Research Program(No.2018Z05JDX005).
文摘Inspired by real biological neural models,Spiking Neural Networks(SNNs)process information with discrete spikes and show great potential for building low-power neural network systems.This paper proposes a hardware implementation of SNN based on Field-Programmable Gate Arrays(FPGA).It features a hybrid updating algorithm,which combines the advantages of existing algorithms to simplify hardware design and improve performance.The proposed design supports up to 16384 neurons and 16.8 million synapses but requires minimal hardware resources and archieves a very low power consumption of 0.477 W.A test platform is built based on the proposed design using a Xilinx FPGA evaluation board,upon which we deploy a classification task on the MNIST dataset.The evaluation results show an accuracy of 97.06%and a frame rate of 161 frames per second.
基金partly supported by the National Natural Science Foundation of China(Nos.62072266 and 62050340)Beijing Academy of Artificial Intelligence(No.BAAI2019ZD0403)。
文摘Brain-inspired computing refers to computational models,methods,and systems,that are mainly inspired by the processing mode or structure of brain.A recent study proposed the concept of"neuromorphic completeness"and the corresponding system hierarchy,which is helpful to determine the capability boundary of brain-inspired computing system and to judge whether hardware and software of brain-inspired computing are compatible with each other.As a position paper,this article analyzes the existing brain-inspired chips design characteristics and the current so-called"general purpose"application development frameworks for brain-inspired computing,as well as introduces the background and the potential of this proposal.Further,some key features of this concept are presented through the comparison with the Turing completeness and approximate computation,and the analyses of the relationship with"general-purpose"brain-inspired computing systems(it means that computing systems can support all computable applications).In the end,a promising technical approach to realize such computing systems is introduced,as well as the on-going research and the work foundation.We believe that this work is conducive to the design of extensible neuromorphic complete hardware-primitives and the corresponding chips.On this basis,it is expected to gradually realize"general purpose"brain-inspired computing system,in order to take into account the functionality completeness and application efficiency.
基金supported by the National Key Research&Development Program of China(No.2016YFB1000504)the National Natural Science Foundation of China(Nos.61877035,61433008,61373145,and 61572280).
文摘Multi-Clock Snapshot Isolation(MCSI)is a concurrency control mechanism that implements snapshot isolation on a single-layer Non-Volatile Memory(NVM)database.It stores a single copy of data by using multi-version storage to ensure durability and runtime access.With multi-clock transaction timestamp assignment,MCSI can efficiently generate snapshots with vector clocks and use per-thread transaction status arrays to identify uncommitted versions in NVM.For evaluation,we compared MCSI with the PostgreSQL-style concurrency control used in the single-layer NVM database N2DB.The maximum transaction throughput of MCSI is 101%–195%higher than that of N2DB for the YCSB workloads,and 25%–49%higher for the TPC-C workloads.Moreover,the transaction latency of MCSI remains relatively stable as the thread count increases.With 18 worker threads,the average transaction latency of MCSI is 65%–84%lower than that of N2DB for the YCSB workloads and 16%–43%lower for the TPC-C workloads.
基金supported by the National Key Research & Development Program of China (No. 2016YFB1000504)the National Natural Science Foundation of China (Nos. 61877035, 61433008, 61373145, and 61572280)。
文摘Non-Volatile Memory(NVM) offers byte-addressability and persistency. Because NVM can be plugged into memory and provide low latency, it offers a new opportunity to build new database systems with a single-layer storage design. A single-layer NVM-Native DataBase(N2 DB) provides zero copy and log freedom. Hence, all data are stored in NVM and there is no extra data duplication and logging during execution. N2 DB avoids complex data synchronization and logging overhead in the two-layer storage design of disk-oriented databases and in-memory databases. Garbage Collection(GC) is critical in such an NVM-based database because memory leaks on NVM are durable. Moreover, data recovery is equally essential to guarantee atomicity, consistency, isolation, and durability properties. Without logging, it is a great challenge for N2 DB to restore data to a consistent state after crashes and recoveries. This paper presents the GC and data recovery mechanisms for N2 DB. Evaluations show that the overall performance of N2 DB is up to 3:6 higher than that of InnoDB. Enabling GC reduces performance by up to 10%,but saves storage space by up to 67%. Moreover, our data recovery requires only 0:2% of the time and half of the storage space of InnoDB.
