R. Jones. The Public Netperf Homepage. http://www.netperf.org/.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....topic areas, including: Kernel microbenchmark suites: the Ousterhout benchmarks [17] lmbench [15] hbench:OS [5] File systems: the Modified Andrew Benchmark (MAB) 17] Chen and Patterson s self scaling benchmark [8] Bonnie, IObench, IOZone. Network throughput and latency: netperf [14], ttcp. Microbenchmarks are widely employed because they are easy to use, they provide simple numeric results (seldom requiring statistical analysis) and because their results are easily correlated with implementation features. Brown and Seltzer have pointed out, however, that creating a ....

R. Jones. The Public Netperf Homepage. http://www.netperf.org/.

....Our reconfiguration time does not include the time to detect and stabilize an inventory change; that takes an additional 15 ms. The detection time depends on the granularity of the FreeBSD clock and could be shortened by making the clock tick faster. 6. 2 Throughput and Latency We used netperf [11] to measure the maximum throughput from one host to another at various packet sizes both with and without a SmartBridge between them. Link and transport overheads are not counted in the throughput. If these overheads are included, our 100 Mb s system runs at full link bandwidth for maximum size ....

Rick Jones. The Public NetPerf Homepage. available at http://www.netperf.org/netperf/NetperfPage.html.

....such buffering since the offered load would originate from significantly more than two machines, and random early drop(RED) could be implemented at the proxy or other router. For traffic generation and performance monitoring we used two applications: netperf, which is publicly available from [6], and loadclient, which we implemented ourselves. Netperf is a network benchmarking tool providing an interface for choosing socket buffer sizes, measuring throughput and response time between two end systems, and recording CPU utilization. Since the netperf client sends all data to the server ....

Rick Jones. The public netperf homepage. Web White Paper. Available as http://www.cup.hp.com/netperf/NetperfPage.html.

....shown in Figure 2. 10 20 30 40 50 60 70 80 90 100 110 120 130 140 1 4 16 64 256 1024 4096 16384 65536 Nachrichtengroesse Theoretische Maximaldurchsaetze Line rate ATM TCP Figure 2: Theoretical achievable TCP bandwidth depending on message size. For measurements, the netperf utility has been used [6]. No particular optimizations involving kernel changes have been made, thus making the results more useful in practice. Only user level adjustable parameters, such as message size and socket options, have been varied. During all measurements the participating machines were otherwise unloaded. 3 ....

Rick Jones. The Public Netperf Homepage. http://www.cup.hp.com/netperf/, 1997.

....would not require such buffering since the offered load would originate from significantly more than two machines, and RED [2] could be implemented at the proxy or other router. For traffic generation and performance monitoring we used two applications: netperf, which is publicly available from [6], and loadclient, which we implemented ourselves. Netperf is a network benchmarking tool providing an interface for choosing socket buffer sizes, measuring throughput and response time between two end systems, and recording CPU utilization. Since the netperf client sends all data to the server ....

Rick Jones. The public netperf homepage. Web White Paper. Available as http://www.cup.hp.com/netperf/NetperfPage.html.

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