Note:The Cisco and Juniper devices were not tested. Typical energy consumption calculation were based on data provided by each vendor on its company website. While no testing was conducted of Cisco or Juniper products, Tolly did offer to share the energy consumption model with each vendor.
Cisco offered corrections to its configuration which are integrated into the published model.
A Juniper representative stated that "We are not endorsing any comparisons not based on actual measurements." The Juniper representative also noted that Juniper supports only two measurement methodologies: ATIS-TEER and ECR
So according to Tolly's test, the Force10 ExaScale configured with the maximum possible 140 10GbE ports consumed ~5,000W less energy than the Cisco Nexus 7000 and ~1,350W less energy than the Juniper EX8216. Outfitted with 10GbE ports, the Force10 ExaScale drew only 3.34W per Gbps of throughput compared with calculations of 7.59W for the Cisco Nexus 7000 and 4.69W for the Juniper EX8216 (see below Figures 1, 2 as well as Table 1):
