Download Question 1: Question 2: Question 3: Question 4:

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Intelligent maintenance system wikipedia , lookup

Prognostics wikipedia , lookup

Transcript
Spring Term 2013
ADVANCED COMPUTER NETWORKS
Assignment 7: Layer7 Switching and High Performance Networking
Assigned on: 25 April 2013
Due by:
9 May 2013
Question 1:
To access services from the data center, we need to locate a “close-by” data center, balance load
across machines in a data center and target machines where the user’s state is kept. Explain which
techniques can be utilized to achieve these three requirements.
Question 2:
The paper: “TCP Splicing for Application Layer Proxy Performance” [1] discusses in detail several
subtleties of implementing the TCP Splice in the proxy. Please explain briefly how the TCP options
are handled in the TCP Splice implementation:
• TCP “urgent data”
• Selective Acknowledgements
• Maximum Segment Size
Question 3:
Read the paper “The Landscape of Computer Architecture” [2] and answer the following questions:
• What trends does Figure 1 show?
• What is the implication of these trends?
Question 4:
Explain briefly what are direct networks and indirect networks, and compare these two types of
networks.
Question 5:
a) Data centers operated as one large layer 2 network allow virtual machines to be migrated
without the need of changing the VMs IP address. Describe why Ethernet as it is does not scale
well to such large deployments?
b) VL2 and PortLand share similar approaches in several aspects in providing large layer-2 networks for data centers. For example, they both use multi-rooted tree topologies. Please describe
two other areas where both methods share similar approaches. Please give succinct answers in the
bullet form. Hint: think about addressing.
c) Identify one potential weakness of each of the two approaches Portland and VL2.
Hand In Instructions
This is a paper exercise. Please hand it in during the exercise session or email to your assistants
by the due date.
References
[1] David A. Maltz and Pravin Bhagwat. “TCP Splicing for Application Layer Proxy Performance”. IBM Research Report 21139 (Computer Science/Mathematics), IBM Research Division, 1998.
[2] “The Landscape of Computer Architecture”, John Shalf, NERSC/LBNL, presented at ISC07,
Dresden, June 25, 2007.
[3] Radhika Niranjan Mysore, Andreas Pamboris, Nathan Farrington, Nelson Huang, Pardis Miri,
Sivasankar Radhakrishnan, Vikram Subramanya, and Amin Vahdat. 2009. “PortLand: a scalable fault-tolerant layer 2 data center network fabric”. SIGCOMM Comput. Commun. Rev.
39, 4 (August 2009), 39-50.
[4] Albert Greenberg, James R. Hamilton, Navendu Jain, Srikanth Kandula, Changhoon Kim,
Parantap Lahiri, David A. Maltz, Parveen Patel, and Sudipta Sengupta. 2009. VL2: a scalable
and flexible data center network. In Proceedings of the ACM SIGCOMM 2009 conference on
Data communication (SIGCOMM ’09). ACM, New York, NY, USA, 51-62.
2