DOC PREVIEW
Purdue CS 59000 - Improving the Scalability of Data Center Networks

This preview shows page 1-2-3 out of 8 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 8 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 8 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 8 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 8 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Slide 1IntroductionIntroductionAdvantagesAdvantagesDisadvantagesDisadvantagesReferenceImproving the Scalability of Data Center Networkswith Traffic-aware Virtual Machine PlacementAuthors:Xiaoqiao Meng, Vasileios Pappas, Li ZhangPresented By:Manoj Raj PenmetchaIntroduction•The paper presents novel approaches to reduce the latencies and increase the efficiency of networks by viewing the placement of VMs within the topology of the datacenter•Apart from the standard improvisations on the network routing algorithms and network topology, the paper present approaches to increase the effective bandwidth between VM pairs with high data transfer and reduce the latency between such pairsIntroduction•Here it uses a 2 Tier approach Algorithm•First it clusters the slots available on the physical machines using the switch distance as the clustering measure and then cluster the VMs using the cost matrix as the clustering measure.•Then the algorithm tries to match the clusters first and eventually the members within the matched clusters.Advantages•Approach is toward the placement of the VMs and not towards the scaling of the network hardware. This is useful where the network hardware is not being currently used to the full potential due to bottlenecks.•The beneficial topology conditions and other network parameters which would lead to useful application of the optimization problem are also studied.Advantages•Apart from the solution to the optimization problem, the effect of traffic models, topology models such as the Global Traffic Model, Partitioned Traffic Model are also presented.Disadvantages•The paper completely ignore the effect of network related optimization on the result of the optimization of placement of VMs.•The min cut algorithm used for clustering the slots available on the physical machines is of high polynomial complexity (O(n^4)) , which is not ideal for large datacenters.Disadvantages•It has been proved that the TVMPP has polynomial time solutions only for the global traffic model, but the real traffic model in a datacenter is a hybrid model varying from the global traffic model, appropriate optimizations for a real scenario have not been aptly discussedReference•Improving the Scalability of Data Center Networks with Traffic-aware Virtual Machine Placement. by Xiaoqiao Meng, Vasileios Pappas, Li Zhang IBM T.J. Watson Research Center 19 Skyline Drive, Hawthorne, NY


View Full Document

Purdue CS 59000 - Improving the Scalability of Data Center Networks

Documents in this Course
Lecture 4

Lecture 4

42 pages

Lecture 6

Lecture 6

38 pages

Load more
Download Improving the Scalability of Data Center Networks
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Improving the Scalability of Data Center Networks and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Improving the Scalability of Data Center Networks 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?