Transcription of Data Center Network Architectures
1 data Center Network ArchitecturesJuha SaloAalto University School of Science and centers have become increasingly essential part of In-ternet communications, thus there have been interest in un-derstanding how to better design and manage data centers[2]. In this paper we explain a typical data Center Network ar-chitecture in the industry, the challenges modern data centernetworks encounter today and introduce proposed solutionsby a recent : data Center Network , cloud computing, opti-mization1 IntroductionAccording to Costa et al. [6], data centers run by Mirosoft,Yahoo, Google and Amazon host tens of thousands serversto provide services across the Internet and the only compo-nent that has not changed during the vast development in datacenters is the data Center Layer 3 Layer 2 Internet BR BR AR AR AR AR AS AS S S S S ToR Servers ToR Servers ToR Servers BR = L3 Border Router AR = L3 Access Router AS = L2 Aggr Switch S = L2 Switch ToR = Top-of-Rack Switch.
2 Figure 1: A typical data Center Network architecture by [9, 8]that is an adaptation of figure by Cisco [5].Typical data Center Network architecture usually consistsof switches and routers in two- or three-level hierarchy [1].Figure 1 is an instance of three-level design. The followingexplanation for the figure is based on the papers [9, 8]. Thefigure contains hierarcy of top layer core routers on the topconnecting to the servers in racks at the bottom. The hierar-chy consists of data Center layer 3, layer 2 and Top of Rackswitches connecting servers.
3 data Center layer 3 includes re-quests arriving from the Internet as IP. Layer 3 access andborder routers manage traffic into and out of the data switches (AS) aggregate connections to layer 3access routers (AR) and provide redundancy. Layer 2 do-main contains usually several thousands servers that are iso-lated to server groups by virtual LAN (VLAN) of Rack (ToR) switch usually connects 20 to 40 serversin the rack by 1 Gbps link. In addition, all links use Ethernetas protocol for Physical Enterprise and Cloud Service data Cen-tersAccording to Greenberg et al.
4 [7], major companies, such asGoogle, Microsoft and Yahoo have made vast scale invest-ments for data centers supporting cloud services. Also by thesame authors [7], data centers supporting cloud services dif-fer from typical enterprise data centers. First, cloud servicedata centers require automation, in contrast to an enterprisedata Center where automation might be just partial. Second,cloud service data centers must support large economies ofscale. For instance, 100 000 servers might be reality. Last,cloud service data centers should scale out, not to out distributes workload to low cost hardware, incontrast to updating high cost hardware.
5 Instances of an en-terprise data Center designs are Ethane [3] and SEATTLE[12].By Greenberg et al. [9], the enterprise networking ar-chitectures were originally developed for much smaller datacenters, in contrast to the ones existing today. The limita-tions of the traditional architecture have resulted in severalworkarounds and patches for the protocols to keep up withthe new demands on data et al. [8] states cloud services relation to datacenters and the problem with agility. Cloud services havebeen a major factor for the innovation of data centers andintroduces agile thinking in the data centers.
6 Cloud ser-vices provide bulk deployments and flexible reallocation ofservers to different services. However, these improvementshave an economic impact. Shared data Center of 100 000servers costs $12 million per month. The current data centerdesigns prevent them being highly profitable and economi-cal, because the designs does not provide mandatory in data Center means responding to different demandsmore rapidly, instead of trying to avoid any change by pre-allocating resources for unkown demands. Agile design indata centers promises cost savings and improved risk Seminar on Internetworking2010-05-05, updated Cost structureIn [7] Greenberg et al.
7 Presents data about the cost structurein a data Center . Networking in data centers consumes 15%of total costs, as shown in table 1. However, networking hasa more widespread impact on the whole system. Innovatingthe networking is the key to reduce the total , memory, storage25%InfrastructurePower distribution, cooling15%Power drawElectrical utility costs15%NetworkLinks, transit, equipmentTable 1: Typical data Center costs [7].By [7], the greatest portion of total costs belong to theservers. To allow efficient use of the hardware, a high levelof utilization, the data centers should provide a method todynamically grow the number of servers and allow focusingresources on optimal locations.
8 Now the fragmentation ofresources prevents the server to Greenberg et al. [7], reducing infrastructurecosts might depend on allowing scale out model of low costservers. Scaling out in a data Center might mean shiftingthe responsibility of expensive qualities of servers, such asfailure rate from a single server to the whole system. Byallowing the Network architecture to scale out, the low failurerate is ensured by having multiple cheap servers, rather thana few expensive [7], power related costs are similar to the Network devices consume 59% of each watt delivered, 8% to dis-tribution losses and 33% for cooling.
9 Cooling related costscould be reduced by allowing the data centers to run hotter,thus maybe requiring the Network to be more resilient Greenberg et al. [7] note that significant fraction ofnetwork related costs goes to networking equipment. Otherportions of the total costs of the Network relate to wide areanetworking, including traffic to end users, traffic betweendata centers and regional facilities. Reducing the networkcosts focuses on optimizing the traffic and data Center this paper we focus only on the problems and solutionsof the data Center Network Architectures .
10 Section 2 of thispaper covers the main problems of Network , section 3 evaluates different proposed solutions. Last,in conclusion we summarize the the main Problems with Network Scalability and physical constraintsThe data Center scaling out means addition of componentsthat are cheap, whereas in scaling up more expensive compo-nents are upgraded and replaced to keep up with demand [9].According to Guo et al. [10], a more attractive performance-to-price ratio can be achieved by using commodity hardware,because the per-port cost is cheaper for the commodity hard-ware than with the more technically advanced onesCabling complexity could be a practical barrier for scal-ability, for instance in modular data Center (MDC) designthe long cables between the containers cause an issue as thenumber of containers increase [14].
