Energy Efficient Tapered Data Networks for Big Data processing in IP/WDM networks

Abstract

Classically the data produced by Big Data applications is transferred through the access and core networks to be processed in data centers where the resulting data is stored. In this work we investigate improving the energy efficiency of transporting Big Data by processing the data in processing nodes of limited processing and storage capacity along its journey through the core network to the data center. The amount of data transported over the core network will be significantly reduced each time the data is processed therefore we refer to such a network as an Energy Efficient Tapered Data Network. The results of a Mixed Integer linear Programming (MILP), developed to optimize the processing of Big Data in the Energy Efficient Tapered Data Networks, show significant reduction in network power consumption up to 76%.

Greening big data networks: Volume impact

“…In this work, we extend our previous work in [9] and [10] to further investigate the impact of the volumes of big data on network power consumption. We optimize the processing location of different amounts of volumes of data generated by different zones of the core networks so that processing could be achieved at the source core node and/or along the data journey through the core network to the datacentres.…”

“…Note that we use processor cycles in GHz as a measure of the total processing capability of a node [11]. Table I summarises the input parameters of the model [8], [9], [12]- [19].…”

“…The concept of greening big data networks is illustrated in Figure 1 [9]. Figugre 1(a) displays the conventional techniuqe of proccessing big data where the Chunks (the raw data before processing with large volume) are transffered from the source to the datacenters for storing and processing.…”

“…Max server power consumption 300 W [8], [9] Energy per bit of the LAN switch 11.875 W/Gbps [12] Energy per bit of the LAN router 7.727 W/Gbps [12] Storage power consumption 0.008 W/Gb [12] IP/WDM router power consumption 825 W [9] , [19] IP/WDM regenerator power consumption 222 W [19] , [9] IP/WDM transponder power consumption 167 W [19] , [9] IP/WDM optical switch power consumption Figure 2. The processing and storage capacities of the PNs are randomly assigned between 10-30 servers (with 4 GHz CPU capacity per server) and 10 PB -75 PB, respectively.…”

Energy Efficient Big Data Networks: Impact of Volume and Variety

“…In [17], we presented preliminary results to demonstrate that the Network Power Consumption (NPC) is affected by processing and transferring big data in bypass IP over WDM networks. We considered one big data type from the MapReduce platform that was obtained from the log files of MapReduce clusters from Facebook [18].…”