^ Top

NANOG Meeting Presentation Abstract

Building Scalable Data Centers: BGP is the Better IGP
Meeting: NANOG55
Date / Time: 2012-06-04 10:30am - 11:00am
This item is webcast
Room: Bayshore Ballroom A-C
Presenters: Speakers:

Petr Lapukhov, Microsoft

Petr Lapukhov is a network engineer at Microsoft Global Networking Services (GNS) Search team. The team is dedicated to supporting Bing Search engine as well as other online services at Microsoft, such as Bing Maps, Ad center, and MSN among others. The primary focus of the team is on areas of Data-Center/WAN network design, implementation and operations. GNS Search works in close cooperation with Autopilot – a team that develops and operates the software to automate Bing’s large-scale data-center infrastructure. Petr has more than 14 years of experience in networking, starting back in 90’s with Novell Netware/IPX, coaxial cables, Cisco 2500 boxes and dialup networks. Prior to joining Microsoft he was working as a CCIE instructor and network engineer at various positions in Russian companies and his alma mater – Kazan State University where he got his MSc in Applied Mathematics.
Abstract: Design choices, such as protocols & topology, for large scale (100k+ services) Web data centers have significant impact on scalability and operations of underlying networks. Such networks often host Hadoop-like compute clusters, which calls for topology providing large amount of bisection bandwidth. Consequently, to satisfy the requirement for intense east-west traffic, operators commonly choose some of Clos/Fat-tree topology variants. Another prominent property of such environments is reliance on fixed form-factor, cheap commodity 1G/10G Ethernet switches to reduce capital expenditures. Both of these properties combined result in a network topology with very high link density and large number of switches – on the order of 10’s of thousands and thousands of elements respectively. Applying traditional data-center designs with large L2 domains, even leveraging features such as vPC/mLAG is not possible due to scalability issues. This situation calls for different design options
Files: pdfBuilding Scalable Data Centers: BGP is the Better IGP(PDF)
youtubeBuilding Scalable Data Centers: BGP is the Better IGP
Sponsors: None.

Back to NANOG55 agenda.

NANOG55 Abstracts

  • MPLS OAM
    Speakers:
    Sam Aldrin, Huawei Technologies;
  • Track: DNS
    Speakers:
    Mehmet Akcin, ICANN; Duane Wessels, Verisign;
  • Track: DNS
    Speakers:
    Mehmet Akcin, ICANN; Duane Wessels, Verisign;

 

^ Back to Top