kumoh national institute of technology
Networked Systems Lab.

Review Comment

NSL > Education> Review Comment
Alif - Computer Standards and Interfaces
By :
Date : 2018-03-20
Views : 35

Ref: CSI_2017_338
Title: Overhead Reduction Scheme for SDN-based Data Center Networks
Journal: Computer Standards & Interfaces
Dear Professor Kim,
Thank you for submitting your manuscript to Computer Standards & Interfaces. We have completed the review of your manuscript. A summary is appended below. While revising the paper please consider the reviewers' comments carefully. We look forward to receiving your detailed response and your revised manuscript.
To submit your revised manuscript:
• Log into EVISE at: http://www.evise.com/evise/faces/pages/navigation/NavController.jspx?JRNL_ACR=CSI
• Locate your manuscript under the header 'My Submissions that need Revisions' on your 'My Author Tasks' view
• Click on 'Agree to Revise'
• Make the required edits
• Click on 'Complete Submission' to approve
What happens next?
After approving your submission you will receive a notification that the submission is complete. To track the status of your paper throughout the editorial process, log into EVISEat: http://www.evise.com/evise/faces/pages/navigation/NavController.jspx?JRNL_ACR=CSI
Enrich your article to present your research with maximum impact. This journal supports the following Content Innovations:
• Explain your research in your own words and attract interest in your work using AudioSlides : 5-minute webcast-style presentations that are displayed next to your published article and can be posted on other websites. You will receive an invitation email to create an AudioSlides presentation within three weeks after your paper has been accepted.
• Interactive Plots: Interactive plot viewer providing easy access to the data behind plots. Please prepare a .CSV file with your plot data and test it online here before submitting as supplementary material.

Data in Brief (optional)

We invite you to convert your supplementary data (or a part of it) into a Data in Brief article. Data in Brief articles are descriptions of the data and associated metadata which are normally buried in supplementary material. They are actively reviewed, curated, formatted, indexed, given a DOI and freely available to all upon publication. Data in Brief should be uploaded with your revised manuscript directly to Computer Standards & Interfaces. If your Computer Standards & Interfaces research article is accepted, your Data in Brief article will automatically be transferred over to our new, fully Open Access journal, Data in Brief, where it will be editorially reviewed and published as a separate data article upon acceptance. The Open Access fee for Data in Brief is $500.

Please just fill in the template found here:
http://www.elsevier.com/inca/publications/misc/dib_data%20article%20template_for%20other%20journals.docx. Then, place all Data in Brief files (whichever supplementary files you would like to include as well as your completed Data in Brief template) into a .zip file and upload this as a Data in Brief item alongside your Computer Standards & Interfaces revised manuscript. Note that only this Data in Brief item will be transferred over to Data in Brief, so ensure all of your relevant Data in Brief documents are zipped into a single file. Also, make sure you change references to supplementary material in your Computer Standards & Interfaces manuscript to reference the Data in Brief article where appropriate.

Questions? Please send your inquiries to dib@elsevier.com. Example Data in Brief can be found here:
http://www.sciencedirect.com/science/journal/23523409
I look forward to receiving your revised manuscript as soon as possible.
Kind regards,
Rory O'Connor
Editor-in-Chief
Computer Standards & Interfaces



Comments from the editors and reviewers:
-Reviewer 1

- The authors provide a framework for SDN-based data center networks in order to reduce overhead between the SDN controller and OF-switches in DCNs. Their main idea is that the SDN architecture model for DCNs is based on the flow-level rather than the packet-level, however, this idea is not new, and may be one of the versions of OpenFlow specification they may provide either packet- or flow-level service. They didn't mention which version of OF specification is used in their implementation. Therefore, the authors should provide more detailed information about using OF specification and what is a novelty of the ideas in the paper clearly.

-Reviewer 2

- Please compare your proposed method with similar methods (Not only the original !)
There are many new methods have been proposed in the last 1.5 years but are not included in the related works.

-Reviewer 3

- This paper proposes some methods for reducing the controller workload and the switch flow table size.
Only the first packet sent to the controller. But will it bring a trouble on synchronization at the next switches among the following packets before the rules updated? To reduce the overhead on the switch by source routing, this idea is similar to some previous work such as JumpFlow.

-Reviewer 4

- This paper proposed a framework on controller and switches to reduce overhead for SDN-based Data Center Networks. The proposed method can boost network performance by reducing the number of PACKET IN and PACKET OUT messages, higher throughput, and lower end-to-end latency.
This paper is appropriate for the journal, and well-organized. Simulation seems to be done considering details. I could find several English grammar errors which are expected to be corrected.

-Reviewer 5

-
This paper considers the scalabilty issues in SDN. Two mechanisms have been proposed to reduce the message overhead between controllers and switches. The first mechanism is to reduce the number of packet-in messages by processing only the first packet of a flow in the controller. The second mechanism is to reduce the rule installation by using a MPLS-like mechanism.
This paper is lack of novelty. For the first mechanism, in fact, most current SDN applications handles packet-in messages just like this! That is, in reactive mode, when a SDN application receives a packet-in message, it calculates forwarding rules, and install rules in the sending switch and other switches in the forwarding path to setup the flow path. Thus, I cannot see any novelty in the first mechanism.
The second mechanism considers how to install minimal flow rules in the case where the packet traverses a long forwarding path. As the author said, it is motivated by [15]. However, [15] has contained the solutions for the scenario that this paper considers, as it is said that Owing to the hop limit, a packets complete routing information must be divided into several sections and loaded at different switches on its forwarding route in [15]. Thus, there is no novelty in this paper.
Some detailed comments:
1. The introduction section should be rephrased. The content is not concise and introduce many things about the background which should be put in Section 3.
2. I dont realize the meaning of section 2.1, I think it should be deleted, for it has nothing to do with your two research aspects.
3. The size of topologies in the evaluation is small and I dont think the results are convicing.
4. Why the traffic rate is set to 10,000 packets/s? I think more experiments should be conducted with different traffic rate.
5. I suggest there should be a paragraph that introduce the content of the whole section at the beginning of each section to make the paper more clear.
6. This sentence is not clear,"This paper proposes a framework on controller and switches to reduce overhead for SDN-based Data Center Networks (DCNs)", reduce what overhead?
7. "Generally, they are inflexible and do not effectively utilize available resources, including CPUs, memory, and storage to run todays network activities", what do they refer to? Its hard to understand.
8."and this first part is reduces the number of PACKET IN message -> is to reduce
9. "To enhance performance, many support techniques to reduce lack of flexibility, scalability, and efficiency have been proposed," This sentence should be rephrased.
10. "This paper discusses the requirements for reducing controller and switch overhead in SDN-based DCN as follows:" This sentence should be rephrased.
11. "the algorithm allocates i as 1 to be the counter of switch set S involved in the flow forwarding path." This sentence is confusing.
12. Ref. [14] is not complete, what is the source?



Have questions or need assistance?
For further assistance, please visit our Customer Support site. Here you can search for solutions on a range of topics, find answers to frequently asked questions, and learn more about EVISE via interactive tutorials. You can also talk 24/5 to our customer support team by phone and 24/7 by live chat and email.
-------------------------------------------------------------
Copyright © 2018 Elsevier B.V. | Privacy Policy
Elsevier B.V., Radarweg 29, 1043 NX Amsterdam, The Netherlands, Reg. No. 33156677.