{"id":25248,"date":"2026-11-17T13:27:59","date_gmt":"2026-11-17T13:27:59","guid":{"rendered":"https:\/\/atalnetworks.com\/?p=25248"},"modified":"2026-11-17T13:27:59","modified_gmt":"2026-11-17T13:27:59","slug":"test-10gbps-server-iperf3","status":"publish","type":"post","link":"https:\/\/atalnetworks.com\/de\/test-10gbps-server-iperf3\/","title":{"rendered":"How to Test a 10Gbps Dedicated Server with iperf3"},"content":{"rendered":"<p><span style=\"font-weight: 400;\">To test a 10Gbps dedicated server with iperf3, connect it to a second capable host and measure TCP throughput in both directions. Compare one stream with four parallel streams, then save the results with the test conditions. This shows how much traffic the selected endpoints and network path can carry during the test.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">A port speed alone cannot tell you how fast a backup, download, or replication job will run. This guide explains how to set up a controlled Linux test, read the output, check unexpected results, and close the temporary listener afterward.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">The 10Gbps Test Plan at a Glance<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">A useful 10Gbps network bandwidth test compares four cases: one outgoing stream, one incoming stream, four outgoing streams, and four incoming streams. Keep the endpoints and settings fixed so the results are comparable.<\/span><\/p>\n<ol>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Prepare two capable hosts and record their starting state.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Start a listener with access limited to the test source.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Measure one TCP stream in each direction.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Compare parallel streams, with optional UDP or full-duplex tests.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Read the receiver results alongside CPU and network counters.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Save a report and check any result that misses your target.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Stop the listener and remove temporary access rules.<\/span><\/li>\n<\/ol>\n<p><span style=\"font-weight: 400;\">Use this matrix as the baseline:<\/span><\/p>\n<table>\n<thead>\n<tr>\n<th><b>Test<\/b><\/th>\n<th><b>Traffic direction<\/b><\/th>\n<th><b>TCP streams<\/b><\/th>\n<th><b>Initial duration<\/b><\/th>\n<th><b>Question answered<\/b><\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td><span style=\"font-weight: 400;\">T1<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Dedicated server to peer<\/span><\/td>\n<td><span style=\"font-weight: 400;\">1<\/span><\/td>\n<td><span style=\"font-weight: 400;\">30 seconds<\/span><\/td>\n<td><span style=\"font-weight: 400;\">How fast is one outgoing flow?<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">T2<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Peer to dedicated server<\/span><\/td>\n<td><span style=\"font-weight: 400;\">1<\/span><\/td>\n<td><span style=\"font-weight: 400;\">30 seconds<\/span><\/td>\n<td><span style=\"font-weight: 400;\">How fast is one incoming flow?<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">T3<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Dedicated server to peer<\/span><\/td>\n<td><span style=\"font-weight: 400;\">4<\/span><\/td>\n<td><span style=\"font-weight: 400;\">30 seconds<\/span><\/td>\n<td><span style=\"font-weight: 400;\">How much outgoing traffic can several flows carry?<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">T4<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Peer to dedicated server<\/span><\/td>\n<td><span style=\"font-weight: 400;\">4<\/span><\/td>\n<td><span style=\"font-weight: 400;\">30 seconds<\/span><\/td>\n<td><span style=\"font-weight: 400;\">How much incoming traffic can several flows carry?<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p><span style=\"font-weight: 400;\">Thirty seconds and four streams are starting choices for this procedure, not universal pass criteria. The commands below are documentation-based examples. No figures in this guide represent a completed Atal Networks hardware benchmark.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">What iperf3 Measures, and What It Leaves Out<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">iperf3 measures traffic between a client and a listening server. A normal test generates data in memory, which helps separate network transfer performance from disk reads and writes. A file copy adds storage and application work that this test does not reproduce.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Keep three terms separate:<\/span><\/p>\n<ul>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Link rate:<\/b><span style=\"font-weight: 400;\"> The capacity reported for a network interface, such as 10Gbps.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Throughput:<\/b><span style=\"font-weight: 400;\"> The transfer rate achieved during a stated test.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Transfer volume:<\/b><span style=\"font-weight: 400;\"> The amount of data moved, such as 50GB.<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">A physical 10Gbps network port is also different from TCP or UDP port 5201. The physical port carries traffic; the numbered transport port directs traffic to the test application.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Choose a Path That Answers Your Question<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">A nearby peer helps establish a baseline with fewer Internet-path variables. A distant peer helps measure delivery to that destination. Neither test isolates every component between the hosts.<\/span><\/p>\n<table>\n<thead>\n<tr>\n<th><b>Test path<\/b><\/th>\n<th><b>Useful for<\/b><\/th>\n<th><b>Main limit<\/b><\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td><span style=\"font-weight: 400;\">Two capable hosts on a nearby network<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Initial throughput baseline<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Does not establish remote Internet performance<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Two hosts in different regions<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Regional transfer planning<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Includes both endpoints and the wider route<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">A public test endpoint<\/span><\/td>\n<td><span style=\"font-weight: 400;\">An additional path comparison<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Capacity and current load may be unknown<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Loopback on one host<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Local software checks<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Does not test the physical NIC or uplink<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p><span style=\"font-weight: 400;\">Do not use <\/span><span style=\"font-weight: 400;\">127.0.0.1<\/span><span style=\"font-weight: 400;\"> to measure a dedicated server&#8217;s external connection. Loopback traffic stays on the host.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Step 1: Prepare Both Hosts Before Generating Traffic<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">Use two authorized hosts with enough spare CPU and network capacity for the rate you want to measure. A slow peer can limit the result even when the dedicated server has spare capacity.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The guide uses consistent roles:<\/span><\/p>\n<ul>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Server A:<\/b><span style=\"font-weight: 400;\"> The dedicated server under test. Run all client commands here.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Server B:<\/b><span style=\"font-weight: 400;\"> The peer. Run the listener here.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>A to B:<\/b><span style=\"font-weight: 400;\"> Traffic leaving the dedicated server.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>B to A:<\/b><span style=\"font-weight: 400;\"> Traffic entering the dedicated server.<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">If B is a VM, check its guest bandwidth and CPU allocation. A host&#8217;s 10Gbps uplink does not establish how much capacity one guest can use.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Install iperf3 and record the Versions<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">On Debian or Ubuntu, install the package on both hosts:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">sudo apt update<\/span><\/p>\n<p><span style=\"font-weight: 400;\">sudo apt install iperf3<\/span><\/p>\n<p><span style=\"font-weight: 400;\">On an RPM-based distribution that provides the package in its enabled repositories:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">sudo dnf install iperf3<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Record the installed version on each host:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">iperf3 &#8211;version<\/span><\/p>\n<p><span style=\"font-weight: 400;\">iperf3 3.16 and later use a separate thread for each test stream. Older builds handle streams differently, so record both versions before comparing results. iperf2 and iperf3 are not compatible endpoints for the same test.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">If installation offers to start a permanent server, decline for this temporary procedure. Preserve any existing service configuration that other users need.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Check the Interface and Starting Load<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">On A, identify the route to B:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">ip route get LISTENER_IP<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Replace <\/span><span style=\"font-weight: 400;\">LISTENER_IP<\/span><span style=\"font-weight: 400;\"> with B&#8217;s reachable IPv4 address. Replace <\/span><span style=\"font-weight: 400;\">INTERFACE<\/span><span style=\"font-weight: 400;\"> below with the interface shown by that route lookup:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">ip -s link show dev INTERFACE<\/span><\/p>\n<p><span style=\"font-weight: 400;\">ethtool INTERFACE<\/span><\/p>\n<p><span style=\"font-weight: 400;\">ethtool -i INTERFACE<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Check B&#8217;s interface as well. Record the reported speed, MTU, driver, and counter values. Some virtual interfaces do not expose physical-link information, and some queries need elevated permissions.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">A reported <\/span><span style=\"font-weight: 400;\">10000Mb\/s<\/span><span style=\"font-weight: 400;\"> confirms only the local link rate when the driver reports it correctly. It does not establish throughput across the route.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Budget for the Whole Test, Not One Command<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">A 10Gbps rate sustained for 30 seconds corresponds to 37.5GB before accounting for how payload and billing are measured:<\/span><\/p>\n<p><b>10,000,000,000 bits\/s \u00d7 30 seconds \u00f7 8 = 37,500,000,000 bytes<\/b><\/p>\n<table>\n<thead>\n<tr>\n<th><b>Test plan<\/b><\/th>\n<th><b>Raw rate-time volume at 10Gbps<\/b><\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td><span style=\"font-weight: 400;\">One 30-second, one-way run<\/span><\/td>\n<td><span style=\"font-weight: 400;\">37.5GB<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Four baseline runs, executed separately<\/span><\/td>\n<td><span style=\"font-weight: 400;\">150GB<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Three repeats of all four baseline runs<\/span><\/td>\n<td><span style=\"font-weight: 400;\">450GB<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p><span style=\"font-weight: 400;\">These calculations assume each run sustains 10Gbps. Actual useful payload can be lower. Retransmissions, framing, warm-up traffic, and billing rules affect other traffic totals. Optional tests add more traffic.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Choose an agreed test window. A full-rate test can consume capacity needed by live applications.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Step 2: Start a Listener with Restricted Access<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">Start the listener on B and allow only the intended source through the active firewall. Keep existing security controls in place while testing.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">For an existing UFW-managed firewall, inspect its current policy:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">sudo ufw status verbose<\/span><\/p>\n<p><span style=\"font-weight: 400;\">If UFW is active, an example source-specific TCP rule is:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">sudo ufw allow proto tcp from CLIENT_IP to LISTENER_IP port <\/span><span style=\"font-weight: 400;\">5201<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Replace <\/span><span style=\"font-weight: 400;\">CLIENT_IP<\/span><span style=\"font-weight: 400;\"> with A&#8217;s source address as B sees it. NAT can change that address. Replace <\/span><span style=\"font-weight: 400;\">LISTENER_IP<\/span><span style=\"font-weight: 400;\"> with B&#8217;s local listening address. UFW supports source, destination, protocol, and port conditions in its rules.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">A narrow allow rule does not cancel an existing broad allow rule. Check the complete policy, provider firewall, and any network access controls. Use your current firewall manager rather than enabling a second one for this test.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">On B, start the listener:<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">iperf3 -s -B LISTENER_IP<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">Leave that terminal open. The default application port is TCP 5201.\u00a0<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Check Connectivity at a Limited Rate<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">On A, run a short check before attempting full throughput:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">iperf3 -c LISTENER_IP -P <\/span><span style=\"font-weight: 400;\">1 <\/span><span style=\"font-weight: 400;\">-b <\/span><span style=\"font-weight: 400;\">100M <\/span><span style=\"font-weight: 400;\">-t <\/span><span style=\"font-weight: 400;\">10<\/span><\/p>\n<p><span style=\"font-weight: 400;\">This requests a 100Mbps TCP rate for 10 seconds. Use a lower target if that load would affect other traffic.<\/span><\/p>\n<p><b>Expected result:<\/b><span style=\"font-weight: 400;\"> The connection opens, data transfers, and the run completes. This checks the setup; it does not establish 10Gbps capacity.<\/span><\/p>\n<table>\n<thead>\n<tr>\n<th><b>Setup problem<\/b><\/th>\n<th><b>First checks<\/b><\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td><span style=\"font-weight: 400;\">Connection refused<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Listener process, address, port, and rejecting rules<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Connection timeout<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Reachability, routing, filtering, and actual source address<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Address already in use<\/span><\/td>\n<td><span style=\"font-weight: 400;\">An existing service on the selected address and port<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Server busy<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Another active test; coordinate before retrying<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Cannot bind address<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Whether the requested local address belongs to B<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p><span style=\"font-weight: 400;\">Resolve setup errors before interpreting performance.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Step 3: Measure a Single TCP Stream in Both Directions<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">Run one stream from A to B, then reverse the data direction from the same client. These tests establish whether one flow behaves differently when sending or receiving.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Create a results directory on A. Choose a new directory name for each test session:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">mkdir -p iperf-results-session-01<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">Run the forward test:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">iperf3 -c LISTENER_IP -P <\/span><span style=\"font-weight: 400;\">1 <\/span><span style=\"font-weight: 400;\">-t <\/span><span style=\"font-weight: 400;\">30 <\/span><span style=\"font-weight: 400;\">| tee iperf-results-session-01\/t1-forward-one.txt<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">Run the reverse test after the first one finishes:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">iperf3 -c LISTENER_IP -P <\/span><span style=\"font-weight: 400;\">1 <\/span><span style=\"font-weight: 400;\">-t <\/span><span style=\"font-weight: 400;\">30 <\/span><span style=\"font-weight: 400;\">-R | tee iperf-results-session-01\/t2-reverse-one.txt<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">The commands display the output and save it. Existing files with the same names are replaced, so use new names for repeat runs. Check that each test completed successfully before accepting its output.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">These baseline commands have no explicit rate cap. Run them only within the agreed capacity and traffic budget.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Keep Direction Labels Clear<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">Reverse mode changes the data direction, not which host starts the test. A initiates both commands. In T1, A sends; in T2, A receives.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Write \u201cA to B\u201d and \u201cB to A\u201d in the report. Labels such as upload and download become confusing when the operator switches between hosts.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Include Startup in the First Baseline<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">The first baseline includes startup behavior. If you later use an omitted warm-up period to study steady-state performance, record that choice and compare it only with matching runs.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Short transfers and long transfers answer different operational questions. A backup may need sustained throughput, while a small download may finish before a long test reaches its steady state.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Step 4: Compare Parallel Streams and Optional UDP Tests<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">Compare four TCP streams with the single-stream baseline to see how the endpoints and path handle several flows. Keep full-duplex and UDP tests separate because they change the workload.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">On A, run four streams in each direction:<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">iperf3 -c LISTENER_IP -P <\/span><span style=\"font-weight: 400;\">4 <\/span><span style=\"font-weight: 400;\">-t <\/span><span style=\"font-weight: 400;\">30 <\/span><span style=\"font-weight: 400;\">| tee iperf-results-session-01\/t3-forward-four.txt<\/span><\/p>\n<p><span style=\"font-weight: 400;\">iperf3 -c LISTENER_IP -P <\/span><span style=\"font-weight: 400;\">4 <\/span><span style=\"font-weight: 400;\">-t <\/span><span style=\"font-weight: 400;\">30 <\/span><span style=\"font-weight: 400;\">-R | tee iperf-results-session-01\/t4-reverse-four.txt<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">Let the first command finish before the second starts. A higher aggregate rate does not mean that every individual connection can reach that rate.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Parallel flows can distribute work differently across CPUs and network paths. Keep the one-stream result even if the parallel result looks better.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Optional: Test Both Directions at Once<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">Use a separate bidirectional run when the application sends and receives heavily at the same time:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">iperf3 -c LISTENER_IP -P <\/span><span style=\"font-weight: 400;\">1 <\/span><span style=\"font-weight: 400;\">-t <\/span><span style=\"font-weight: 400;\">30 <\/span><span style=\"font-weight: 400;\">&#8211;bidir<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Check support on both builds. Record incoming and outgoing rates separately. Adding both rates does not produce a one-way throughput figure.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Optional: Measure UDP Loss and Jitter<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">UDP testing helps compare offered traffic with received traffic under a specified load. It is useful for workloads where packet delivery and timing variation matter.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Keep TCP access for the control connection and permit UDP test traffic. If the earlier UFW example applies to B:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">sudo ufw allow proto udp from CLIENT_IP to LISTENER_IP port <\/span><span style=\"font-weight: 400;\">5201<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Start with a bounded rate on A:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">iperf3 -c LISTENER_IP -u -P <\/span><span style=\"font-weight: 400;\">1 <\/span><span style=\"font-weight: 400;\">-b <\/span><span style=\"font-weight: 400;\">100M <\/span><span style=\"font-weight: 400;\">-l <\/span><span style=\"font-weight: 400;\">1200 <\/span><span style=\"font-weight: 400;\">-t <\/span><span style=\"font-weight: 400;\">10<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The example requests 100Mbps with a 1,200-byte UDP payload. Confirm that the payload plus headers fits the path MTU, including any tunnel overhead. Increase the offered rate only within the agreed budget and after reviewing endpoint load.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Keep the first UDP comparison to one stream. With parallel streams, the bitrate target applies to each stream. Record requested rate, actual sender and receiver rates, payload size, lost datagrams, and jitter.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">A requested rate is not proof that the host generated or delivered that rate. Jitter describes timing variation; it does not replace a website response-time test.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Step 5: Read the Receiver Rate Alongside Host Load<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">Use the final receiver summary for received throughput, then check intervals, retransmissions, and host load for context. A peak interval alone cannot describe the completed run.<\/span><\/p>\n<table>\n<thead>\n<tr>\n<th><b>Field<\/b><\/th>\n<th><b>What it tells you<\/b><\/th>\n<th><b>Reading rule<\/b><\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td><span style=\"font-weight: 400;\">Interval<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Time covered by the row<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Separate whole-run results from short intervals<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Transfer<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Data counted over that interval<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Check the displayed unit<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Bitrate<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Rate over the interval<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Do not select only the highest row<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Sender\/receiver<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Endpoint perspective<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Match the role to the test direction<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">SUM<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Combined streams<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Compare with another aggregate result<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Retr<\/span><\/td>\n<td><span style=\"font-weight: 400;\">TCP retransmissions where available<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Review with load and counter changes<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Cwnd<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Congestion-window information where available<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Treat as TCP context, not link capacity<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p><span style=\"font-weight: 400;\">Available fields vary by protocol, platform, and build. Keep the complete output rather than copying only the largest number.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Do Not Treat Sender-Receiver Differences as a Loss Formula<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">TCP sender and receiver summaries can differ because their measurement boundaries and data in flight differ. Subtracting one rate from the other does not calculate packet loss.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Retransmissions also need context. A count without the transfer duration, traffic volume, and path conditions cannot identify a failed component.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Check Both Hosts During the Run<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">A single busy CPU core can limit a transfer while average server CPU use looks low. If the <\/span><span style=\"font-weight: 400;\">sysstat<\/span><span style=\"font-weight: 400;\"> tools are installed, run this in a separate session on each host:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">mpstat -P ALL <\/span><span style=\"font-weight: 400;\">1 35<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Compare interface counters before and after the same run:<\/span><\/p>\n<p><span style=\"font-weight: 400;\">ip -s link show dev INTERFACE<\/span><\/p>\n<p><span style=\"font-weight: 400;\">An increasing drop counter deserves investigation, but its meaning depends on the interface and driver. Do not assume every drop is a physical fault.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Keep the first baseline on the existing configuration. Changes to buffers, offloads, or congestion control belong in separate comparisons with recorded starting values and a recovery plan.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Step 6: Decide Whether the Result Meets Your Requirement<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">Judge a test against a target that names the path, direction, stream count, and workload. A 10Gbps port does not create one universal acceptance threshold for all Internet destinations.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">10Gbps equals 1,250MB\/s before overhead in decimal units. Ethernet, IP, transport headers, and other processing reduce useful payload. MB\/s and MiB\/s also use different byte units.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Die <\/span><a href=\"https:\/\/atalnetworks.com\/de\/10gbps-vs-1gbps-dedicated-servers\/\"><span style=\"font-weight: 400;\">10Gbps vs. 1Gbps dedicated server comparison<\/span><\/a><span style=\"font-weight: 400;\"> explains the capacity difference. For this test, agree on the required received rate and permitted variation before judging the result.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Use Repeat Runs Without Hiding Variation<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">If the traffic budget permits, repeat each baseline case three times. Report the median and minimum-to-maximum range for each case. Three repeats are a practical first check, not evidence of performance throughout a day or month.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Keep different destinations, IP families, stream counts, and directions in separate rows.<\/span><\/p>\n<p>&nbsp;<\/p>\n<table>\n<thead>\n<tr>\n<th><b>Test<\/b><\/th>\n<th><b>Direction<\/b><\/th>\n<th><b>Streams<\/b><\/th>\n<th><b>Median receiver rate<\/b><\/th>\n<th><b>Min\u2013max range<\/b><\/th>\n<th><b>CPU \/ counter notes<\/b><\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td><span style=\"font-weight: 400;\">T1<\/span><\/td>\n<td><span style=\"font-weight: 400;\">A to B<\/span><\/td>\n<td><span style=\"font-weight: 400;\">1<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Record result<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Record range<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Record observations<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">T2<\/span><\/td>\n<td><span style=\"font-weight: 400;\">B to A<\/span><\/td>\n<td><span style=\"font-weight: 400;\">1<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Record result<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Record range<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Record observations<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">T3<\/span><\/td>\n<td><span style=\"font-weight: 400;\">A to B<\/span><\/td>\n<td><span style=\"font-weight: 400;\">4<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Record result<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Record range<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Record observations<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">T4<\/span><\/td>\n<td><span style=\"font-weight: 400;\">B to A<\/span><\/td>\n<td><span style=\"font-weight: 400;\">4<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Record result<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Record range<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Record observations<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p><span style=\"font-weight: 400;\">Preserve OS, kernel, iperf3 version, CPU, NIC, driver, link rate, MTU, addresses, locations, exact command, timestamp, and background load with the table.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Save JSON When You Need Machine-Readable Results<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">Add JSON output for a separately recorded run:<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">iperf3 -c LISTENER_IP -P <\/span><span style=\"font-weight: 400;\">1 <\/span><span style=\"font-weight: 400;\">-t <\/span><span style=\"font-weight: 400;\">30 <\/span><span style=\"font-weight: 400;\">-J &gt; iperf-results-session-01\/t1-forward-one-json-run.json<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">The usual JSON document is completed at the end of the test. Review errors and interrupted runs before using them in a summary. This extra command creates another transfer, so include it in the traffic budget.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Follow the Result Pattern to the Next Check<\/span><\/h3>\n<table>\n<thead>\n<tr>\n<th><b>Pattern<\/b><\/th>\n<th><b>What to check next<\/b><\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td><span style=\"font-weight: 400;\">One stream is slow; four are faster<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Per-core CPU use, TCP behavior, and flow distribution<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Only one direction is slow<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Both host roles, directional policies, and routing<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Nearby is fast; remote is slow<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Remote endpoint capacity, latency, and route<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">CPU reaches a limit during low throughput<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Host processing capacity and tool version<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">Retransmissions or drops rise with load<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Endpoint counters, queues, and possible path loss<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">iperf3 is fast; file copies are slow<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Storage, encryption, application settings, and differing traffic policies<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p><span style=\"font-weight: 400;\">A second authorized endpoint can help determine whether a result follows one peer or path. Change one variable at a time.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">For example, a backup job may need fast disk writes at B even when B receives memory-based test traffic well. Repeat the actual backup task before deciding that the service meets its operational target.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Step 7: Close the Listener and Restore Temporary Rules<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">Stop the test process and remove only the access rules added for this procedure. Keep the evidence files so future tests have a baseline.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">On B, press <\/span><span style=\"font-weight: 400;\">Ctrl+C<\/span><span style=\"font-weight: 400;\"> in the listener terminal. If you created the UFW rules shown earlier, remove them with their matching descriptions:<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">sudo ufw delete allow proto tcp from CLIENT_IP to LISTENER_IP port <\/span><span style=\"font-weight: 400;\">5201<\/span><\/p>\n<p><span style=\"font-weight: 400;\">sudo ufw delete allow proto udp from CLIENT_IP to LISTENER_IP port <\/span><span style=\"font-weight: 400;\">5201<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">Remove the UDP rule only if you added it. Do not remove pre-existing service access. UFW supports deletion by the original rule description.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Check for a remaining TCP listener:<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">ss -ltnp <\/span><span style=\"font-weight: 400;\">&#8216;sport = :5201&#8217;<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">If the test was the only service on that port, no matching row is expected. Identify any remaining process before stopping it.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">On a systemd host, check whether an iperf3 service exists:<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">systemctl status iperf3<\/span><\/p>\n<p>&nbsp;<\/p>\n<p><span style=\"font-weight: 400;\">A missing unit can be normal. Stop and disable a service only if this test created it and no other work requires it. Restore any other settings changed during diagnosis.<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">H\u00e4ufig gestellte Fragen<\/span><\/h2>\n<h3><span style=\"font-weight: 400;\">Can one TCP stream fill a 10Gbps connection?<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">One TCP stream can achieve high throughput when the hosts and path support it. CPU processing, TCP behavior, latency, and loss affect the result. Compare one stream with several flows to see whether a single-flow limit matters to your workload. Do not discard the single-stream result just because the aggregate number is higher.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Do I need a second 10Gbps dedicated server?<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">You need a peer and path capable of the rate you want to measure. The peer does not need the same server model, but it needs enough available network and processing capacity. A slower peer can measure lower-rate performance; it cannot establish the first server&#8217;s capacity against a capable 10Gbps endpoint.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Can I use a public iperf3 server?<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">A public endpoint can provide an additional comparison if its operator permits the test. Check usage rules, ports, rate limits, and allowed duration first. Its current load may be unknown. Treat the result as performance to that endpoint, and use controlled peers when you need a repeatable acceptance check.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Does iperf3 measure disk speed?<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">The normal commands in this guide generate memory-based traffic and do not reproduce disk reads or writes. A file copy adds storage, filesystem, and application work. Use iperf3 to establish the network baseline, then run a representative file transfer to check the complete task. Separate those results in your report.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">What is the difference between reverse and bidirectional tests?<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">A reverse test sends data from the listener back to the client. A bidirectional test sends data both ways at once. The client starts either test, but the load differs. Record each direction independently and compare bidirectional results only with other runs that use the same combined load.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Why does a UDP test also need TCP access?<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">iperf3 uses a TCP control connection to coordinate a UDP test. The test data uses UDP, so both paths need the required access through the relevant firewalls. Allow the specific test source and destination rather than opening the listener broadly, and remove temporary access when the test finishes.\u00a0<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">How do I test IPv6 with iperf3?<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">Use the peer&#8217;s reachable IPv6 address and select IPv6 with the <\/span><span style=\"font-weight: 400;\">-6<\/span><span style=\"font-weight: 400;\"> option. Confirm listener binding and IPv6 firewall access on that path. Keep IPv6 and IPv4 results separate because the address families may follow different routes or policies. Record the family beside each test result so later comparisons remain clear.<\/span><\/p>\n<h3><span style=\"font-weight: 400;\">Can I compare Windows and Linux iperf3 results?<\/span><\/h3>\n<p><span style=\"font-weight: 400;\">Compare them only after recording the exact builds and networking environments. ESnet does not officially support native Windows iperf3, and third-party builds can differ. A WSL test also includes a virtual networking layer. Use comparable Linux endpoints for the baseline when possible, then measure the actual client environment separately.\u00a0<\/span><\/p>\n<h2><span style=\"font-weight: 400;\">Recommended Next Steps<\/span><\/h2>\n<p><span style=\"font-weight: 400;\">Keep the four-case baseline and use it to check the real workload. If a backup, download, or replication job still misses its target, review application and storage behavior alongside the network results.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">For an unexplained network result, send support the raw logs, timestamps, endpoint details, commands, and host observations. A specific, repeatable test gives the team a useful starting point.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">If the workload needs more capacity, use the <\/span><a href=\"https:\/\/atalnetworks.com\/de\/how-to-choose-a-10gbps-dedicated-server\/\"><span style=\"font-weight: 400;\">10Gbps server selection guide<\/span><\/a><span style=\"font-weight: 400;\"> to review hardware and network requirements. Then compare <\/span><a href=\"https:\/\/atalnetworks.com\/de\/10gbps-server\/\"><span style=\"font-weight: 400;\">Atal Networks 10Gbps dedicated servers<\/span><\/a><span style=\"font-weight: 400;\"> and share the measured requirements with our team.<\/span><\/p>","protected":false},"excerpt":{"rendered":"<p>To test a 10Gbps dedicated server with iperf3, connect it to a second capable host and measure TCP throughput in [&hellip;]<\/p>\n","protected":false},"author":2,"featured_media":25249,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"default","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"set","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[1],"tags":[],"class_list":["post-25248","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-enterprise-grade-server"],"acf":[],"_links":{"self":[{"href":"https:\/\/atalnetworks.com\/de\/wp-json\/wp\/v2\/posts\/25248","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/atalnetworks.com\/de\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/atalnetworks.com\/de\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/atalnetworks.com\/de\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/atalnetworks.com\/de\/wp-json\/wp\/v2\/comments?post=25248"}],"version-history":[{"count":1,"href":"https:\/\/atalnetworks.com\/de\/wp-json\/wp\/v2\/posts\/25248\/revisions"}],"predecessor-version":[{"id":25250,"href":"https:\/\/atalnetworks.com\/de\/wp-json\/wp\/v2\/posts\/25248\/revisions\/25250"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/atalnetworks.com\/de\/wp-json\/wp\/v2\/media\/25249"}],"wp:attachment":[{"href":"https:\/\/atalnetworks.com\/de\/wp-json\/wp\/v2\/media?parent=25248"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/atalnetworks.com\/de\/wp-json\/wp\/v2\/categories?post=25248"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/atalnetworks.com\/de\/wp-json\/wp\/v2\/tags?post=25248"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}