文章对吉首市"湘西黄金茶"区域公共品牌建设的相关做法和经验进行梳理和总结,并就如何进一步高效推进"湘西黄金茶"区域公共品牌建设提出了相关对策,旨在为吉首市茶产业发展献计献策,助力吉首茶农增收致富.

茶树修剪是茶园优质高产的关键技术措施.通过对湖南省吉首市茶园修剪作业中存在的主要问题进行梳理和总结,就如何把握合适的修剪方式、选择恰当的修剪时机、合理控制留养叶、科学跟进剪后培管措施等提出了具体改进措施,并开展了建设性思考,旨在为吉首茶农增产增收提供技术参考.

整合大数据流提高道路通行效率 复工复产,交通运输是"先行官".在云南省公安厅交警总队合成作战指挥中心,全省各主要路口、路段,重点车辆、驾驶人,旅游景区、商业网点道路数据流源源不断汇入集成指挥平台,为指挥调度提供最可靠的参数指标.通过大数据分析比对与高清视频巡检结合,指挥中心以整体管控、缩减周期、提高效率为原则,对道路交通的管控方案进行适时调整,保持全省联勤联动,严格执行 24 小时勤务模式,为疫情期间企业复工复产打开"绿色通道".

<正>2012年9月7日11时19分,云南省昭通市彝良县与贵州省毕节地区威宁彝族回族苗族自治县交界发生5.7级地震。地震经专家认定为"震群型"、"浅源型"地震,震级虽不大,但破坏力很大,据不完全统计,地震造成了昭通市范围共81人死亡,800多人受伤,上万间房屋倒塌,数万间房屋受损,人民群众生命财产受到严重损失。

To assure the trustworthiness of high-trusted architecture of embedded operating system that named Hades,the inter-partition information flow of Hades must be controlled strictly.By analyzing the limitation of existing methods,we present a new information flow control model and construct a information flow mechanism between partitions.The accuracy of these model and mechanism are verified through the experiment.

Due to the ignorance of energy and security factors,traditional scheduling policies are not suitable for security-critical distributed embedded systems.A Security and Energy-Aware Task Mapping and Scheduling(SEATMS) algorithm is proposed for security and energy-critical distributed collaborative applications.The proposed algorithm can strive to reduce the security risk while without sacrificing the real-time and energy constraints,and was proved to be an algorithm with polynomial time complexity.Simulation results show superiority of SEATMS on real-time guarantee,energy budget guarantee,and low security risk.

A recursive bipartitioning algorithm,RPM,is proposed for low energy mapping in tree-based Network-on-Chip (NoC) architectures subject to communication latency constraints. The mapping problem is formulated to multi-level IP core communication task graph partitioning problems,and the modified Kernighan-Lin mincut heuristic is used to solve them. Experimental results show that RPM obtains lower energy mapping solutions compared with existing algorithms.

Real-time garbage collection should collect unused memory and guarantee real-time tasks to meet their deadlines. Furthermore,system memory requirement should be considered. A hybrid garbage collection is proposed in this paper. Reference counting algorithm is used to collect acyclic garbage and timestamp algorithm is used to collect cyclic garbage. GC and real-time tasks are scheduled concurrently. The proposed GC not only collects all unused memory but also reduces memory requirement compared with hybrid GC based on mark-sweep algorithm. It fits for large-scale system.

Dynamic memory management with garbage collection is very important for development of large-scale software system.To apply garbage collection into embedded real-time system,real-time garbage collection is researched in this paper.Dynamic execution of garbage collection and scheduling garbage collection by EDF are proposed.Memory analysis and simulation results show that proposed algorithms can reduce memory requirement and enhance flexibility of system compared with previous work.Distributed garbage collection is also researched and performance analysis shows that proposed distributed garbage collection based on critical reference verification fits for large-scale distributed system.

Without considering security, existing message scheduling mechanisms may expose critical messages to malicious threats like confidentiality attacks. Incorporating confidentiality improvement into message scheduling, this paper investigates the problem of scheduling aperiodc messages with time-critical and security-critical requirements. A risk-based security profit model is built to quantify the security quality of messages; and a dynamic programming based approximation algorithm is proposed to schedule aperiodic messages with guaranteed security performance. Experimental results illustrate the efficiency and effectiveness of the proposed algorithm.

To improve the dependability of security/safety-critical systems,after analyzing status quo of high dependable safeguard mechanism,a multi-layered architecture based on the concept of separation and the VM is proposed,which supports applications with multi-level security.This paper researches multi-layered security/safety policy,the information flow control mechanism and the evaluation and certification for trusted software.This architecture can provide trustworthy services for the embedded security/safety-critical systems.

High quality of security and guaranteed real-time requirements are two key goals of mission- critical embedded storage systems. But most existing real-time disk scheduling algorithms do not consider improving security performance of disk requests. A security-aware periodic-write (SAPW) scheduling algorithm is proposed to judiciously select appropriate security level for each disk request to maximize security value of N periodic disk users, while without sacrificing timing constraint of each user. Simulation results show the significant effectiveness of SAPW algorithm, and the average security improve- ment is up to 223.6% over other three algorithms.

The problem of energy-aware mapping for network-on-chip(NoC) implemented with voltage islands under reliability constraints is formulated.A tabu search algorithm is proposed to solve this problem.A novel energy-efficiency gradient driven heuristic is developed to assign the voltages for the voltage islands based on the current configuration of IP mapping during global design space exploration.Experimental results show that the proposed method can efficiently guarantee the reliability goal and obtain significant energy savings.

High-assurance embedded systems have very strict requirement in both design and analysis. This paper presents a trusted architecture named Hades to high-assurance system design for security and safety-critical embedded systems with MLS, which is based on the concept of separation. We define a rule that only authorized data can be transferred among partitions to absolutely guarantee the security and safety for high-assurance embedded systems. The inter-partition information flow control mechanism for Hades architecture can prevent untrustworthy and low-level partitions from violating other partitions, by means of which the security and safety of high-assurance embedded systems are assured.

To enhance the dependability and security of complex embedded security safety-critical systems with multi-levels security (MLS),a trusted architecture,called Hades,is presented. Hades is based on the concept of temporal and spatial isolation and the partition mechanism. Each partition shares the physical resource and uses only a fraction of the time,and so it is free from the interference of one another. This paper also proposes a scheduling model of the bounded-delay partition to solve the scheduling problem of partitions,and designs a two-level scheduling mechanism with priority bitmap algorithm using two scheduling policies,Rate Monotonic (RM) and Earliest Deadline First (EDF).The schedulable conditions for preemptive fixed priority and dynamic priority scheduling policies are provided separately. Simulation results demonstrated the effectiveness of the proposed model.

Security-critical systems tend to be subjected to much more serious security threats,therefore it poses a great challenge to reduce the risk of security-critical applications.This article proposes a risk-driven task scheduling mechanism by incorporating risk assessment with real time scheduling.Based on the concept of synthetic utilization of aperiodic tasks,an online task scheduling algorithm is designed to schedule security-sensitive aperiodic tasks with minimal security risk.This algorithm is an approximate algorithm,which can yield security performance guaranteed near-optimal solution in low time complexity without violating the security-critical and time-critical constraints of tasks.Simulation results show that the proposed algorithm can significantly reduce the risk of security-critical real time applications and satisfy the bound of risk approximation ratio.Specifically,maximal security risk reduction of up to 22.2% is achieved,and minimal approximation ratio of security risk can decrease to 102.4% for the bound of 110%.

Security is of critical importance for real-time applications running on pervasive computing environments.Security-critical task model,security overhead model and security risk model are built for real-time periodic tasks,and a security risk-aware adaptive scheduling algorithm is proposed.According to inherent properties of real-time periodic tasks,the scheduling problem is transformed to a multi-stage decision-making procedure with the purpose of minimizing security risk.Based on approximate dynamic programming policy,the proposed algorithm is designed with security performance guaranteed and low-complexity mechanisms.Experimental results show that the proposed algorithm can significantly reduce security risk of applications,satisfy the security requirements,and adapt itself to dynamic pervasive computing environments.

This paper proposed a new garbage collection mechanism,that was real-time distributed garbage collection based on verification of critical reference.Performance analysis showed that the algorithm could identify and collect cyclic garbage at minimal delay compared with previous work.Though the proposed algorithm required more memory space for reference listing and verification,it was real-time and fault tolerance.Thus,the algorithm has high synthetic performance and fits for large-scale distributed system.

A routing protocol based on gradient and cluster in wireless sensor network is analyzed. An energy-efficient gradient-based clustering routing protocol (GBCRP) is proposed and evaluated. The new protocol combines the advantages of gradient and cluster. It saves the energy and improves the reliability by the technologies of clustering, creating gradient based on the hop, key nodes transmitting and the adaptive power optimal routing of cluster header nodes. Simulation results show that the new routing protocol provides longer lifetime and higher reliability.

For decreasing memory requirement of system with real-time garbage collection (GC), a real-time hybrid garbage collection, which is combined with partial mark-sweep algorithm and reference counting algorithm is proposed. Acyclic garbage is collected by reference counting algorithm incrementally and cyclic garbage is also collected by partial mark-sweep algorithm incrementally. Thus, partial cyclic garbage can be collected during one GC cycle for reuse. Hybrid GC can be scheduled with real-time tasks concurrently to make real-time tasks meet their deadlines. Theoretical analysis and experimental results demonstrated superiority in memory consumption of the proposed algorithm in comparison with the existing algorithms.