Automatically configuring the network layer of data centers for cloud computing

Hu, Chengchen; Yang, Mao; Zheng, Kai; Chen, K.; Zhang, Xuan; Liu, B.; Guan, Xiaohong

doi:10.1147/jrd.2011.2165751

Cited by 6 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The recent work [8] also proposes a method DCZeroconf for automatic address configuration in data center networks. DCZeroconf aims to the goal of proposing a fully automatic address configuration method even need no blueprint as input.…”

Section: Related Workmentioning

confidence: 99%

GA-MAP: An Error Tolerant Address Mapping Method in Data Center Networks Based on Improved Genetic Algorithm

Deng¹,

Wang²,

Zhang³

et al. 2015

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARY Address configuration is a key problem in data center networks. The core issue of automatic address configuration is assigning logical addresses to the physical network according to a blueprint, namely logical-to-device ID mapping, which can be formulated as a graph isomorphic problem and is hard. Recently years, some work has been proposed for this problem, such as DAC and ETAC. DAC adopts a sub-graph isomorphic algorithm. By leveraging the structure characteristic of data center network, DAC can finish the mapping process quickly when there is no malfunction. However, in the presence of any malfunctions, DAC need human effort to correct these malfunctions and thus is time-consuming. ETAC improves on DAC and can finish mapping even in the presence of malfunctions. However, ETAC also suffers from some robustness and efficiency problems. In this paper, we present GA-MAP, a data center networks address mapping algorithm based on genetic algorithm. By intelligently leveraging the structure characteristic of data center networks and the global search characteristic of genetic algorithm, GA-MAP can solve the address mapping problem quickly. Moreover, GA-MAP can even finish address mapping when physical network involved in malfunctions, making it more robust than ETAC. We evaluate GA-MAP via extensive simulation in several of aspects, including computation time, error-tolerance, convergence characteristic and the influence of population size. The simulation results demonstrate that GA-MAP is effective for data center addresses mapping.

show abstract

Section: Related Workmentioning

confidence: 99%

GA-MAP: An Error Tolerant Address Mapping Method in Data Center Networks Based on Improved Genetic Algorithm

Deng¹,

Wang²,

Zhang³

et al. 2015

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…[2] proposes DAC, a generic and automatic Data center Address Configuration system, and [6] proposes DCZeroconf, a fully automatic IP address configuration mechanism to eliminate the burden of manual configurations of IP addresses of servers and switches in data centers. These approaches are very useful for large scale data centers that have many servers or devices, but do not consider settings of components like middleware or software running on management servers that are necessary for managing data centers.…”

Section: Related Workmentioning

confidence: 99%

Automatic server role identification for cloud infrastructure construction

Kitajima

Uchiumi

Kikuchi

et al. 2013

2013 IEEE 2nd International Conference on Cloud Networking (CloudNet)

View full text Add to dashboard Cite

The recent progress in computer performance and the development of virtualization technologies has led to the prevalence of cloud computing. Data center providers providing public cloud services have to install additional resources and infrastructures continuously to keep up with the increasing demands from cloud users. Since the newly installed infrastructure (e.g. servers) usually have similar structure as the existing infrastructure, the configuration settings for the existing ones can be copied and used for the new one. One of the exceptions is network setting (e.g. IP address) which must be customized for each infrastructure. However, the customization requires manual configuration, which can cause misconfigurations, resulting in communication failures in the new infrastructure. One of the promising approaches to identify the misconfigurations is to detect the differences between the communication logs recorded in the existing infrastructure and the new infrastructure being developed. In order to execute this approach, we need to identify a pair of servers that play the same role in the existing and new infrastructure so that we can verify whether or not the same functions are working properly in both of these infrastructures. In this paper, we propose a method that automatically identifies the pair of servers playing the same role by detecting the common communication patterns observed in both infrastructures. We evaluated our method in actual cloud infrastructure and confirmed that it identified 94.1% of corresponding pairs of servers correctly.

show abstract

“…PortLand [11] and DCZeroconf [9] also consider address configuration in the context of data center networking. PortLand introduces a distributed location discovery protocol (LDP) for its PMAC address assignment.…”

Section: Related Workmentioning

confidence: 99%

Error Tolerant Address Configuration for Data Center Networks with Malfunctioning Devices

Chen

et al. 2012

2012 IEEE 32nd International Conference on Distributed Computing Systems

View full text Add to dashboard Cite

Abstract-Address auto-configuration is a key problem in data center networks, where servers and switches encode topology information into their addresses for routing. A recent work DAC [2] has been introduced to address this problem. Without malfunctions, DAC can auto-configure all the devices quickly. But in case of malfunctions, DAC requires significant human efforts to correct malfunctions and it can cause substantial operation delay of the whole data center.In this paper, we further optimize address auto-configuration process even in the presence of malfunctions. Instead of waiting for all the malfunctions to be corrected, we could first configure the devices that are not involved in malfunctions and let them work first. This idea can be translated to considerable practical benefits because in most cases malfunctions in data centers only account for a very small portion. To realize the idea, we conceptually remove the malfunctions from the physical data center topology graph and mathematically convert the address configuration problem into induced subgraph isomorphism problem, which is NP-complete. We then introduce an algorithm that can solve the induced subgraph isomorphism quickly by taking advantage of data center topology characteristics and induced subgraph properties. We extensively evaluate our design on representative data center structures with various malfunction scenarios. The evaluation results demonstrate that the proposed framework and algorithm are efficient and labor-free to deal with the mapping task in the presence of error devices.

show abstract

Automatically configuring the network layer of data centers for cloud computing

Cited by 6 publications

References 13 publications

GA-MAP: An Error Tolerant Address Mapping Method in Data Center Networks Based on Improved Genetic Algorithm

GA-MAP: An Error Tolerant Address Mapping Method in Data Center Networks Based on Improved Genetic Algorithm

Automatic server role identification for cloud infrastructure construction

Error Tolerant Address Configuration for Data Center Networks with Malfunctioning Devices

Contact Info

Product

Resources

About