Abstract:With the development of cloud computing technology, Serverless is becoming increasingly popular among developers as an emerging paradigm for building applications in the cloud. Serverless allows developers to focus on the logic of applications without worrying about underlying server management. However, the performance of serverless applications is easily affected when facing cold start issues. In order to mitigate the impact of cold start issues on serverless applications, this paper proposes a cold start mitigation approach for serverless applications based on adaptive container pools to reduce the overall latency caused by cold start issues. This approach optimizes the cold start latency of serverless applications based on the internal structure of serverless applications, through the analysis of container pool mechanisms and request history. Experiments show that this approach has achieved significant effects in improving the performance of serverless applications and the efficiency of cloud resource utilization.

Alert Name:
Alerting to:
Authentication email will be sent to your email address in 24 hours
Frequency:
Email Message Format:	Plain text Graphical(HTML)

Complete the form below and we will recommend the selected titles to your friends on your behalf. * Indicates a required field.
Your name*:
Your email address*:
Recipient's name*:
Recipient's email address*:
(multiple recipient's names and email addresses should be separated with semicolons)
Your comments:	I thought you would find the page(s) useful.

Your name:
Your email address:
Recipient's name:
Recipient's email address:
(multiple recipient's names and email addresses should be separated with semicolons)
Your comments:	I thought you would find this page useful.

Disclaimer: This message was sent to your friend using the "Send it to a friend" facility on the Sciencepaper Online’ WWW site, http://www.paper.edu.cn/en. The Sciencepaper Online is not responsible for the content of this email, and anything said in this email does not necessarily reflect the Sciencepaper Online's views.


	1. Cold Start Mitigation Approach for Serverless Applications Based on Adaptive Container Pool
	LI Zhuo,WANG Chun
	Computer Science and Technology 08 March 2024
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:With the development of cloud computing technology, Serverless is becoming increasingly popular among developers as an emerging paradigm for building applications in the cloud. Serverless allows developers to focus on the logic of applications without worrying about underlying server management. However, the performance of serverless applications is easily affected when facing cold start issues. In order to mitigate the impact of cold start issues on serverless applications, this paper proposes a cold start mitigation approach for serverless applications based on adaptive container pools to reduce the overall latency caused by cold start issues. This approach optimizes the cold start latency of serverless applications based on the internal structure of serverless applications, through the analysis of container pool mechanisms and request history. Experiments show that this approach has achieved significant effects in improving the performance of serverless applications and the efficiency of cloud resource utilization.
	TO cite this article:LI Zhuo,WANG Chun. Cold Start Mitigation Approach for Serverless Applications Based on Adaptive Container Pool[OL].[ 8 March 2024] http://en.paper.edu.cn/en_releasepaper/content/4762581


	2. Efficient Bare Metal Auto-Scaling for NFV Platform
	Pang Xudong,Wang Jing,Wang Jingyu,QiQi
	Computer Science and Technology 19 November 2017
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:There are always a cloud data center behind the Network Function Virtualisation(NFV) system.Typically, elasticity is an essential attribute of cloud data center, which is critical for operating resources in face of peaks and valleys of business. At present, the automatic scaling technique of virtual machines is widely studied, but barely for physical machines. Despite lack of flexibility, we all know that physical server can perform faster and more efficiently than virtualized instances, especially in NFV systems. Some virtual network functions (VNFs) actually require high performance computing, which is a hard task for virtual machines. Besides, good management of bare metal resources can be significant for the data center power cost and human maintenance cost. Accordingly, we think that auto-scaling of physical machine is worth studying. This paper proposes a bare metal automatic scaling scheme based on workload prediction, and finally make tests on an open source NFV platform.The new scheme obtains good result on computation intensive VNFs scenario, including complete the scale in half an hour, guarantee for the continuity of VNF processing business, and can cope with the load fluctuation better.
	TO cite this article:Pang Xudong,Wang Jing,Wang Jingyu, et al. Efficient Bare Metal Auto-Scaling for NFV Platform[OL].[19 November 2017] http://en.paper.edu.cn/en_releasepaper/content/4742224

Saved Papers

Saved Papers


	3. Optimization on Reliability and Memory Usage in Dynamic Stream Processing Engine
	TIAN Jun,LI Jinglin
	Computer Science and Technology 14 November 2017
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:A critical problem in stream processing Engine among multi-nodes is the balance between reliability and low latency. Higher reliability may result in greater latency, while lower latency makes the system more vulnerable. Some application requires strict reliability, meanwhile, system is desired to be more rapid in responsing to messages. Most stream processing engines rely on distributed architecture which is implemented with hashtable to maintain information of distributed nodes. Thus leads to issue in memory occupation. To balance the reliability and latency in SPE and to shrink memory space wasted by nodes, this paper designed a stream processing engine based on ZeroMQ library. This engine enhances reliability by deploying an timing-window algorithm with AOF in Redis, and reduces memory usage in managing information of nodes through double splay tree.
	TO cite this article:TIAN Jun,LI Jinglin. Optimization on Reliability and Memory Usage in Dynamic Stream Processing Engine[OL].[14 November 2017] http://en.paper.edu.cn/en_releasepaper/content/4741957


	4. Low Cost Data Gathering using Mobile Hybrid Sensor Networks
	TAO Dan,TANG Shaojie,MA Huadong
	Computer Science and Technology 05 June 2012
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:In this work we study energy efficient hybrid sensor network design using mobile sinks, motivated by the practical GreenOrbs system application. In our model, the movement of mobile sinks is constrained to be on some predefined road-segments. Two different network structures are investigated: the one-hop structure in which each static sensor can be reached by the mobile sink at some stage of the movement, and the multihop structure where some sensors need the relay by other sensors to reach the sink. The challenge is to find a movement schedule of mobile sink that will minimize the energy cost while meet other constraints. In this work, we first show that the problem is NP-hard and then design an efficient movement scheme and theoretically prove that the total cost is within a constant factor of the optimum. We further present a scheduling solution using integer program for multi-hop structure, which is near optimal and can be computed in polynomial time. Finally, we conduct extensive study of our method in a real wireless sensor network deployment composed of hundreds of static sensors. Our experiments validate the theoretical findings of our method.
	TO cite this article:TAO Dan,TANG Shaojie,MA Huadong. Low Cost Data Gathering using Mobile Hybrid Sensor Networks[OL].[ 5 June 2012] http://en.paper.edu.cn/en_releasepaper/content/4480580


	5. Bandwidth Reduction Appliance in IP Network Remote Mirroring System
	TAN Lejuan,YAO Wenbin
	Computer Science and Technology 23 December 2011
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:This paper designs a bandwidth reduction allpiance in IP network remote mirroring system. For the purpose of transparency, an iSCSI-based redundancy elimination approach combining the variable-sized chunking method and the sliding window method is proposed in the bandwidth reduction allpiance. Firstly, the similarity detection technique, a combination of content-defined chunking (CDC) with Bloom filter, is introduced to find the most similar data block, making redundancy elimination more flexible and efficient. Secondly, a modification is supplied to rsync to make it applicable for iSCSI data blocks, which ensures that the redundancy elimination is transparent to both the local disk and the remote disk. Extensive experiments on ASCII and binary files demonstrate that the new system can achieve a higher performance on synchronization time and bandwidth cost compared to the system without redundancy elimination.
	TO cite this article:TAN Lejuan,YAO Wenbin. Bandwidth Reduction Appliance in IP Network Remote Mirroring System[OL].[23 December 2011] http://en.paper.edu.cn/en_releasepaper/content/4457027


	6. K-means Clustering Distributed Data Mining Based on Hadoop
	XIAO Shan,YU Yanhua,YANG Zhifang,E Xinhua
	Computer Science and Technology 04 December 2011
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract: In the face of massive data on the Internet, a single host has been unable to meet their storage and computing requirements, distributed storage and distributed computing applications is an inevitable trend. Hadoop is the application of which more distributed storage and computing frameworks. Various features of the Hadoop is ideal for data mining, this paper presents a Hadoop-based framework for distributed data mining model, and give the corresponding Map / Reduce process of design ideas and examples for distributed data mining.
	TO cite this article:XIAO Shan,YU Yanhua,YANG Zhifang, et al. K-means Clustering Distributed Data Mining Based on Hadoop[OL].[ 4 December 2011] http://en.paper.edu.cn/en_releasepaper/content/4453607


	7. A Clustered PTT experimental system based on 3G wireless Network
	Wu Honggang,Zhang Honggang,Chai Lunshao,Li Qiaohong
	Computer Science and Technology 29 September 2011
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:This paper discusses the development of the 3G wireless network and a Clustered PTT system based on it, some new features are also introduced into the system. Current communication applications are mostly based on text conversation, voice and video information are hardly real-time because of the limitation of the 2G wireless network. The 3G wireless network support more bandwidth than the 2G's, this let the real-time voice and video communcaitons on wireless data network possible. This paper discuss the structure of a real-time communcations systeam which can support real-time voice and video. The system is also open, other applications can also be added onto it.
	TO cite this article:Wu Honggang,Zhang Honggang,Chai Lunshao, et al. A Clustered PTT experimental system based on 3G wireless Network[OL].[29 September 2011] http://en.paper.edu.cn/en_releasepaper/content/4444033


	8. Extending BPEL for Stateful Service in Convergence Network
	Liu Chen,Zou Hua
	Computer Science and Technology 04 January 2010
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:Network convergence is the trend of future networks to experience various services in a single network. To create and compose services rapidly, variously and efficiently is the main purpose of the service creation in convergence network. BPEL is an important tool in SOA to compose various services in a code- and platform-independent way. In this paper, we discuss the deficiencies of BPEL in modeling stateful telecom service, and then offer an extending proposal for BPEL to support state pattern explicitly to facilitate the stateful service creation in use of BPEL. With the extensions, the service designer could mode a stateful service into a stateful BPEL process, and then the process can be executed on an extended BPEL engine. We also implement a visual editor and an engine based on open-source projects to support our extensions. With the visual editor, the stateful service could be modeled in state pattern visually. And the stateful process can be deployed and executed in the engine.
	TO cite this article:Liu Chen,Zou Hua. Extending BPEL for Stateful Service in Convergence Network[OL].[ 4 January 2010] http://en.paper.edu.cn/en_releasepaper/content/38393


	9. Large-scale Data Analysis based on Hadoop
	Liu Yijun
	Computer Science and Technology 15 December 2009
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:With the development of complex network and the increase of data scale, the performance of data analysis becomes more and more important. In this paper, we present a new approach for network analysis about the flight data based on Hadoop which is an implementation of the MapReduce parallel framework: Firstly, we identity and group the information of passengers and flights. Secondly, we extract graph nodes, which represent the passengers, and graph edges, which represent the relationship between the passengers. Finally we visualize the passenger’s egocentric network to help network analysis.
	TO cite this article:Liu Yijun. Large-scale Data Analysis based on Hadoop[OL].[15 December 2009] http://en.paper.edu.cn/en_releasepaper/content/37600


	10. The Design of PQRM-based Pegasus Workflow System
	Liu Qi ,Zhang Lin
	Computer Science and Technology 08 December 2008
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:Nowadays the workflow application is getting more and more popular to facilitate the grid user to solve complicated computing problems in the distributed resources. Due to the functional mismatch of the workflow management and resource management in the grid computing, we propose a new architecture of PQRM-based Pegasus workflow system, which is able to satisfy the grid workflow builder with SLA guarantees. The functionalities of each component in the integrated system will be specified in this paper as well as the algorithm realizing the SLA specification in the integrated model. Experiments are conducted to evaluate the performance under different SLAs and the discussion based on the trials will be followed at last.
	TO cite this article:Liu Qi ,Zhang Lin. The Design of PQRM-based Pegasus Workflow System[OL].[ 8 December 2008] http://en.paper.edu.cn/en_releasepaper/content/26389

	Check out RSS, or use RSS reader to subscribe this item