Fine-tunning ECN and RTT based Congestion Control in Data Centers Networks Test phase, srs, design phase and source code final deliverable

Get help with
Phd, Masters Thesis & Mcs Final Project
MBA Final Project
Cs619,Fin619,Mgt619,Bnk619,Hrm619,Mkt619
WhatsApp: 0092-3125718857
Skype: trust_aware
Email: projecthelp77@gmail.com
Click here to Join Our Facebook Page
Click here to Join Our YouTube Channel!

Fine-tunning ECN and RTT based Congestion Control in Data Centers Networks Test phase, srs, design phase and source code final deliverable

Project Domain / Category
Networking

Abstract / Introduction
Web programs are gradually moving into cloud environments. These cloud offerings are hosted on huge scale computation and storage infrastructures referred to as data centers (DC) (e.G. Google’s facts center, Facebook’s facts center etc.). In a big scale records middle, loads of heaps of servers are interconnected thru switches in a symmetric topology. A statistics center community (DCN) interconnects all the statistics center sources.
Current facts middle congestion manipulate schemes may additionally induce high latency in packet shipping because of course’s latency-oblivious congestion signal. Explicit congestion notification (ECN) is the predominantly used congestion signal within the information middle networks; it alerts whether any queue alongside the path is above a predefined threshold or not, however does no longer inform approximately the give up-to-cease put off of the course. Information of the cease-to-quit delay / round-journey-time (RTT) of a course can allow a sender to regulate its sending fee to keep the network latency below a threshold. RTTs in facts middle networks are on the size of up to few loads of microseconds and traditional information center operating systems lack such first-class-grained microsecond-degree timers; that is why put off-based totally congestion manipulate schemes, which are widely deployed in the Internet, have no longer been used inside the data center networks. But recent studies like [1] recommend that advances in NIC hardware has enabled correct microsecond-level RTT size, therefore, RTT can be integrated in the facts middle congestion manage schemes.

This mission aims to combine ECN and RTT alerts inside the information center congestion manage scheme and high-quality-tune it. First of all, RTT size (at the source) could be implemented; then the present ECN may be blended with the RTT in the congestion manipulate scheme; finally the congestion manipulate algorithm (primarily based on the ECN and RTT) will be first-rate-tunned to present the ideal outcomes. All implementation may be done in network simulator ns-2. The college students will: i) have a look at an current extensively-deployed DCN congestion manage scheme and its implementation in ns-2, ii) layout changes in the current algorithm(s) to include put off/RTT, iii) put into effect the layout within the current congestion manage scheme in ns-2.

Working in ns-2 calls for: i) expertise of basic commands of Linux running structures (for ns-2 installation and strolling functions), ii) fantastic programming talents in C++ (for simulating the DCN surroundings and implementing the proposed adjustments in the current congestion manipulate scheme), iii) programming in TCL (for writing simulation scripts), iv) expertise of AWK command (for hint text processing), and v) understanding GNUPLOT command (for drawing graphs).

References:
[1] Mittal, R., Lam, V. T., Dukkipati, N., Blem, E., Wassel, H., Ghobadi, M., … & Zats, D. (2015). TIMELY: RTT-based Congestion Control for the Datacenter. ACM SIGCOMM Computer Communication Review, forty five(four), 537-550.

Intended Outcome
• Design of a route’s latency-conscious DCN congestion manage scheme
• Implementation of a direction’s latency-aware DCN congestion control scheme in ns-2

Required Skills
1. Understanding of primary Linux instructions
2. C++, TCL, AWK and GNUPLOT (for working in ns-2)

Supervisor:
Name: Hasnain Ahmed

Leave a Reply

Your email address will not be published. Required fields are marked *

× WhatsApp Us