Computers are taking up great part in technology evolution. They perform many tasks quickly and reliably when compared to humans, opening new avenues. One important aspect of computers is multitasking that enables a s...
详细信息
Real-time systems are intensively used nowadays. scheduling algorithms are very important to manage the scheduling of tasks in real-time systems. In this paper we give an overview on the real-time scheduling technique...
详细信息
We presented a novel platform dedicated to stream processing that improved resource efficiency by sharing resources among applications. The platform utilized latency-aware schedulers to handle stream applications with...
详细信息
Industrial Internet-of-Things, automation, and process control applications require communication networks with a high degree of reliability, low energy expenditure, and low latency. Convergecast communication wherein...
详细信息
ISBN:
(数字)9781728191270
ISBN:
(纸本)9781728191287
Industrial Internet-of-Things, automation, and process control applications require communication networks with a high degree of reliability, low energy expenditure, and low latency. Convergecast communication wherein all nodes in the network transmit data to a common base station is a common paradigm for such applications. Time slotted networks which support deterministic scheduling, such as 802.15.4e Time Slotted Channel Hopping (TSCH) multiple access (MAC) protocol, are well suited for such applications. In this paper, we consider a joint relay selection and scheduling problem for convergecast on a multihop wireless network using single channel TSCH MAC protocol. The objective of the problem is to reduce both the number of relay nodes required for network connectivity as well as the slot-frame duration for a periodic deterministic convergecast schedule. For tree networks with half-duplex constraints we obtain a lower bound on the minimum slot-frame duration and show that the relay selection and scheduling problem can be decoupled. We propose a heuristic scheduling algorithm which is shown to achieve the above lower bound using simulations. For the NP-hard relay selection problem, we propose a polynomial time relay selection algorithm with an improved approximation ratio. The performance of our relay selection and scheduling algorithms are studied via simulations on randomly generated graphs.
In this paper, we are trying to find a algorithm for scheduling DAG (Directed Acyclic Graph) tasks in heterogeneous embedded systems to minimize energy consumption while meeting the reliability requirement. Like many ...
详细信息
In this paper, we are trying to find a algorithm for scheduling DAG (Directed Acyclic Graph) tasks in heterogeneous embedded systems to minimize energy consumption while meeting the reliability requirement. Like many traditional algorithms, we divide the task scheduling algorithm into two phases, the task priority calculation phase and the task allocation phase. In the task priority calculation phase, we proposes a priority calculation algorithm IOD based on the difference in task’s input and output data. In the task allocation stage, we proposes a task allocation algorithm based on fault-tolerant technology of task replication and DVFS technology. Combining the two phase, we get three scheduling algorithms, IODS, IODQ and IODR. In the experimental part, we compare the performance of the algorithm proposed in this paper with existing research algorithms (EFSRG algorithm and HRRM algorithm). The analysis of experimental results shows that the IODS algorithm is a better choice.
The research object of integrated scheduling Institute is single piece and small batch order production. This kind of order is characterized by multiple varieties and small batch, which puts forward a severe test to t...
详细信息
ISBN:
(纸本)9781665400121
The research object of integrated scheduling Institute is single piece and small batch order production. This kind of order is characterized by multiple varieties and small batch, which puts forward a severe test to the flexible production capacity of enterprises. Nowadays, with the rise of Internet of things technology, collaborative cooperation among enterprises is possible. Distributed collaborative manufacturing can effectively integrate enterprise resources, reduce the risks caused by enterprises fighting alone, improve the flexible production capacity of enterprises, and finally enable manufacturing enterprises to improve production efficiency and reduce unknown risks. However, the research on distributed integrated scheduling problem is still in its infancy, and there are few research results. This paper analyzes and studies the research status of the problem from four aspects: symmetric resource distributed integrated scheduling problem, asymmetric resource distributed integrated scheduling problem, multi-objective distributed integrated scheduling problem and special distributed integrated scheduling problem, points out some directions and contents of distributed integrated scheduling optimization algorithm to be further studied, and puts forward suggestions for further research.
Parallel sorting algorithms for manycore CPU systems are needed in data science and big data era. The PDPSort (Parallel Dual Pivot STLSort) can be revised and extended to achieve higher and more stable Speedups. This ...
详细信息
As demand for Large Language Models (LLMs) and AI agents rapidly grows, optimizing systems for efficient LLM inference becomes critical. While significant efforts have targeted system-level engineering, little is expl...
详细信息
Internet is indispensable in all of the activities of today's society. Cloud computing emerges as a demanded recent technology to make the use of internet in a simplified, cost-affordable and timely available form...
详细信息
ISBN:
(纸本)9781665416658
Internet is indispensable in all of the activities of today's society. Cloud computing emerges as a demanded recent technology to make the use of internet in a simplified, cost-affordable and timely available form to the user. The primary purpose of cloud computing is to utilize the resource efficiently, allocate the tasks to the resources, balance their load and satisfy the agreements of user without fail. This research paper includes a brief analysis and comparison of the traditional task scheduling algorithms like FCFS, RR, and MCT with recent algorithms such as IWD, TSCSA, LBMPSO. All of the recent time algorithms are satisfying multi-objective qualities and ensures economic feasibility.
In 5G, network slicing is a prominent feature to provide end-to-end Quality of Experience (QoE) for specific user equipment (UE) requirements. A key technology that facilitates the enhancement of network slicing perfo...
详细信息
ISBN:
(数字)9781665448932
ISBN:
(纸本)9781665448949
In 5G, network slicing is a prominent feature to provide end-to-end Quality of Experience (QoE) for specific user equipment (UE) requirements. A key technology that facilitates the enhancement of network slicing performance is Open-Radio Access Network (ORAN). The support of network slicing is an ongoing technical effort in ORAN. This paves the way for implementing challenging and predictive system control strategies to serve several UEs, leveraging UE specific aggregated data, medium access control (MAC) key performance indicators for optimizing per-UE QoE, service allocation and resource scheduling. In this paper, to improve system performance at a granular level, we propose a novel concept of formation of multiple optimized UE service specific slice-in-slice categories across network slices. There have been no clear state-of-the-art that have addressed the relationships and supportability of UEs of different services across slice-in-slice categories during service allocation in network slicing. In this paper, we propose novel Service Allocation scheduling (SAS) algorithms for UEs services. Novel multi-objective multi-constraint optimization models are formulated to achieve improved service level guarantees. A graph theoretical optimum service allocation strategy is proposed for UE services to maximize the network slice capacity. A probabilistic system resource prediction algorithm for UE services conditioned on previously occurred services is further proposed. Finally, we compared our proposed SAS with state-of-the-art service scheduling algorithms, where SAS ensures 5.12 and 6.22 times throughput gains.
暂无评论