Christian Kurmann. Switchbench benchmark, June 2001. http://www.cs.inf.ethz.ch/CoPs/sbench/. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Cost/Performance Tradeoffs in Network Interconnects for.. - Kurmann, Rauch, Stricker (2003) Self-citation (Kurmann) (Correct)
....or unexptected dependencies on some topological features in the network. In most cases the claim of the vendor that the delivered switch is a full crossbar does not hold and our Switchbench measurement tool shows clearly to what extent this is true or not. The source code can be downloaded from [9]. 9 10] 512 1024 1536 2048 2560 3072 3584 4096 # of routes at a given throughput (9 10] 8 16 24 32 40 48 56 64 # of phases at a given throughput 0 2.5 5 7.5 10 12.5 Overall throughput [MByte s] E7 64x64x40 MB moved 610 s exec. time 4.20 MByte s per ....
Christian Kurmann. Switchbench benchmark, June 2001. http://www.cs.inf.ethz.ch/CoPs/sbench/.
Cost/Performance Tradeoffs in Network Interconnects for.. - Kurmann, Rauch, Stricker (2003) Self-citation (Kurmann) (Correct)
....under AAPC load. Detailed timing data of the different communication phases on all the nodes is gathered to allow careful analysis of the capabilities of the underlying networking architecture or switching technology. The all to all benchmark is integrated into the switchbench benchmark package [10]. 5 Analyzing a Non Performing Ethernet Switch 5.1 Different Network Configurations As announced in Section 1 and described in Section 3 we have three networks installed in the Xibalba cluster: A full bisection primary data network implemented by a large central Ethernet switch, a secondary ....
Christian Kurmann. Switchbench benchmark. http://www.cs.inf.ethz.ch/CoPs/sbench/, June 2001.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC