An energy efficient and runtime-aware framework for distributed stream computing systems

Sun, Dawei; Cui, Yijing; Wu, Minghui; Gao, Shang; Buyya, Rajkumar

doi:10.1016/j.future.2022.06.007

Cited by 5 publications

(3 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It consists two modules: the first module uses Laplacian to partition the topology graph, while the second calculates the node selection method. Sun et al 33 proposed an energy efficient and runtime‐aware framework. They modeled the stream application, resources, and energy consumption, formalizing the scheduling problem.…”

Section: Related Workmentioning

confidence: 99%

Lc‐Stream: An elastic scheduling strategy with latency constraints in geo‐distributed stream computing environments

Sun,

Wang,

Sui

et al. 2024

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

SummaryAn effective scheduling strategy is critical for achieving better performance in real‐time stream processing systems. How to quickly and efficiently process real‐time data stream is always challenging, especially when clusters are collaborating in a Geo‐Distributed computing environment. To address these challenges, we propose an elastic scheduling strategy with Latency Constraints in Geo‐Distributed stream computing environments called Lc‐Stream. This article discusses our work from the following aspects: (1) An optimized data stream redirection method that is proposed based on queuing network algorithm, along with a computing resource model, a latency constrained scheduling model and a communication energy consumption model. (2) An updated node selection method based on the inter‐layer task correlation, to reduce the communication latency between groups at the executor granularity. (3) A network cluster distribution for Geo‐Distributed computing environment to ensure energy saving under low transmission latency. Experimental results show that compared to R‐Storm, Lc‐Stream reduces total latency by over 19% and increases throughput by over 37% in typical cross‐domain multi‐task topologies. Compared to Ts‐Stream, Lc‐Stream also reduces total latency by over 15% and increases throughput by over 21%. At the same time, it helps to balance the load among the systems and avoid overuse of compute nodes.

show abstract

Section: Related Workmentioning

confidence: 99%

Lc‐Stream: An elastic scheduling strategy with latency constraints in geo‐distributed stream computing environments

Sun,

Wang,

Sui

et al. 2024

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

show abstract

“…But there was no discussion on how to make distributed systems have high traffic carrying capacity and stable continuous operation under high concurrency traffic conditions outside the system. Sun et al [21] proposed energy-saving and runtime aware frameworks to improve resource scheduling speed, throughput, latency, and energy consumption. Shen et al [22] utilized node motion trends and mobile centroids to construct adjustable basic positions, and proposed a routing protocol based on mobile centroids, which improved packet delivery rate and reduced latency.…”

Section: Related Workmentioning

confidence: 99%

Traffic Carrying and Delay Response Scheduling Algorithm for Distributed E- commerce Platforms

Gao,

Yang,

Chen

et al. 2024

Preprint

View full text Add to dashboard Cite

The popularization of digitalization, informatization and the Internet has given birth to the rapid development of e-commerce. Faced with the rapidly expanding user traffic, there are still technical bottlenecks in how e-commerce platforms can carry more user traffic and improve server response performance. This article conducts system optimization performance analysis from both hardware and software aspects, and constructs a high-performance distributed AR-AFSA system. (1) The AR (Application Router, AR) architecture is configured with three JobManager server nodes, each receiving three types of user access requests. A traffic allocation mechanism is used to distribute the system's traffic carrying pressure, and user requests are divided into four traffic queues for scheduling according to different access methods. (2) Improve AFSA for container scheduling, re plan the execution order of various behaviors of artificial fish, reduce ineffective search steps, and influence the direction of artificial fish's movement through the global optimal solution, increasing the possibility of finding the optimal solution and accelerating local convergence speed. (3) Using the CPU, memory performance, and load balancing parameters of the container as the parameters and evaluation indicators for artificial fish, matching sufficient resource containers for user requests while ensuring container resource conservation and system load balancing. Finally, the traffic carrying capacity of the AR system and the single JobManager system was validated using the Taobao user behavior dataset and multiple control experiments. The AR system can withstand three times the traffic pressure of traditional servers. The improved AFSA algorithm can converge to a more optimal solution compared to the control algorithm, and in more complex server resource sizes, it consumes lower latency, reduces iteration times, schedules and uses more reasonable resources, demonstrating greater advantages.

show abstract

“…Increasing [6,7,18,21,30,[32][33][34][35] Wave [1, 6, 18, 24, 30-32, 36, 37] Binary [1,5,6,18,19,[22][23][24] Spike [32,[37][38][39] Table 1: Data stream frequency patterns found in the literature.…”

Section: Data Frequency Strategy Related Workmentioning

confidence: 99%

Micro-batch and data frequency for stream processing on multi-cores

et al. 2023

View full text Add to dashboard Cite

Latency or throughput are often critical performance metrics in stream processing. Applications' performance can fluctuate depending on the input stream. This unpredictability is tied to variations in data arrival frequency, data size, complexity, and other factors. Researchers are constantly investigating new ways to mitigate the impact of these variations on performance with self-adaptive techniques involving elasticity or micro-batching. However, there is a lack of benchmarks capable of creating test scenarios to further evaluate these techniques. This work extends and improves the SPBench benchmarking framework to support dynamic micro-batching and data stream frequency management. We also propose a set of algorithms that generate the most commonly used frequency patterns for benchmarking stream processing in related work. It allows the creation of a wide variety of test scenarios. To validate our solution, we use SPBench to create custom benchmarks and evaluate the impact of micro-batching and data stream Micro-batch and Data Frequency for Stream Processing on Multi-cores frequency on the performance of Intel TBB and FastFlow. These are two libraries that leverage stream parallelism for multi-core architectures. Our results demonstrated that most test cases benefited from micro-batches, especially high throughput applications with ordering constraints. For different data stream frequency configurations, TBB ensured the lowest latency, while FastFlow assured higher throughput in shorter pipelines.

show abstract

An energy efficient and runtime-aware framework for distributed stream computing systems

Cited by 5 publications

References 30 publications

Lc‐Stream: An elastic scheduling strategy with latency constraints in geo‐distributed stream computing environments

Lc‐Stream: An elastic scheduling strategy with latency constraints in geo‐distributed stream computing environments

Traffic Carrying and Delay Response Scheduling Algorithm for Distributed E- commerce Platforms

Micro-batch and data frequency for stream processing on multi-cores

Contact Info

Product

Resources

About