Loading

Network Performance Comparison of Fat-Tree and BCube Data Center Architecture: Case Study on Government Office Network
Indra Zulardi1, Nico Surantha2

1Indra Zulardi*, Department of Computer Science, BINUS Graduate Program-Master of Computer Science, Bina Nusantara University Jakarta, Indonesia.
2Nico Surantha, Department of Computer Science, BINUS Graduate Program-Master of Computer Science, Bina Nusantara University Jakarta, Indonesia.
Manuscript received on August 12, 2020. | Revised Manuscript received on August 25, 2020. | Manuscript published on September 10, 2020. | PP: 193-198 | Volume-9 Issue-11, September 2020 | Retrieval Number: 100.1/ijitee.K77620991120 | DOI: 10.35940/ijitee.K7762.0991120
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: The internet now plays crucial roles such as streaming videos, social networking, managing large scientific data. There is a need for high-speed Internet. The volume of data handled by the data centers is increasing very rapidly. Hence, the need to handle data center traffic effectively becomes inevitable. At present, the government office network uses a data center network design based on Fat-Tree that employs high-end IP switches. Unfortunately, the resulting network performance has delays of more than 0.06 ms, throughput of about 500 kbps, and packet loss rate is 2. This research introduces BCube to re-design and improve network performance (i.e., delay, throughput, and packet loss). Through this research, we propose a new data center design in the government office that improves the network performance significantly, with delays less than 0.004 ms, throughput more than 515 kbps, and packet loss rate 0, two–three percent better compared to the Fat-Tree data center network design that contributes significantly to the data center operations in the government office and its overall business performance through the data center network design. 
Keywords: Data Center Networks Design, Network Performance, Delay, Throughput, Packet loss.
Scope of the Article: Software Design Patterns