520 likes | 612 Vues
Learn about tools like AMP, Thrulay, Traceroute for network performance testing. Understand connectivity tests, advanced user tools, and NDT results analysis.
E N D
Tools Overview Richard Carlson January 30, 2007
Basic Premise • Application’s performance should meet your expectations! • If they don’t you should complain!
Underlying Assumption • When problems exist, it’s the networks fault!
Simple Network Picture Bob’s Host Network Infrastructure Carol’s Host
Switch 2 Switch 3 R5 R4 R8 R1 R3 R6 Switch 1 R9 R2 R7 Switch 4 Network Infrastructure
Tools, Tools, Tools • AMP • Advisor • Thrulay • Web100 • MonaLisa • pathchar • NPAD • Pathdiag • Surveyor • Ethereal • CoralReef • MRTG • Skitter • Cflowd • Cricket • Net100 • Ping • Traceroute • Iperf • Tcpdump • Tcptrace • BWCTL • NDT • OWAMP
Basic Connectivity Tests • Ping • Confirms that remote host is ‘up’ • Some network operators block these packets • Traceroute • Identifies the routers along the path • Same blocking problem as above • Routers treat TR packets with lower priority
“ping” results • Intro message • Identifies remote host name and IP address • States size of packets being sent • Setting larger sizes may reveal hidden problems • Output lines • Who responded, and the RTT, maybe other details • Summary lines • Number of packets sent/received/lost • RTT statistics min/average/max Note: 1 msec RTT = 50 miles of between hosts
“traceroute” results • Intro messages • Name and address of remote host • Maximum number of link before giving up • Status messages • One line per router in path • ‘*’ indicates router didn’t respond • Routers usually rate limit replies • No name indicates DNS entry is missing • Hops required to reach remote host or max number from above
Advanced user tools • Existing NDT tool • Allows users to test network path for a limited number of common problems • Existing NPAD tool • Allows users to test local network infrastructure while simulating a long path
Network Diagnostic Tool (NDT) • Measure performance to users desktop • Identify real problems for real users • Network infrastructure is the problem • Host tuning issues are the problem • Make tool simple to use and understand • Make tool useful for users and network administrators
NDT user interface • Web-based JAVA applet allows testing from any browser • Command-line client allows testing from remote login shell
Finding Results of Interest • Duplex Mismatch • This is a serious error and nothing will work right. Reported on main page, on Statistics page, and mismatch: on More Details page • Packet Arrival Order • Inferred value based on TCP operation. Reported on Statistics page, (with loss statistics) and order: value on More Details page
Finding Results of Interest • Packet Loss Rates • Calculated value based on TCP operation. Reported on Statistics page, (with out-of-order statistics) and loss: value on More Details page • Path Bottleneck Capacity • Measured value based on TCP operation. Reported on main page
Finding a Server • What? You don’t have one running at your site? • Install the Internet2 Network Performance Toolkit Knoppix Disk
NPAD/pathdiag • A new tool from researchers at Pittsburgh Supercomputer Center • Finds problems that affect long network paths • Uses Web100-enhanced Linux based server • Web based Java client
Long Path Problem • E2E application performance is dependant on distance between hosts • Full size frame time at 100 Mbps • Frame = 1500 Bytes • Time = 0.12 msec • In flight for 1 msec RTT = 8 packets • In flight for 70 msec RTT = 583 packets
Switch 2 Switch 3 R5 R4 R8 R1 R3 R6 Switch 1 R9 R2 R7 Switch 4 Long Path Problem H2 70 msec H1 – H3 1 msec H1 – H2 X H3 H1
TCP Congestion Avoidance • Cut number of packets by ½ • Increase by 1 per RTT • LAN (RTT=1msec) • In flight changes to 4 packets • Time to increase back to 8 is 4msec • WAN (RTT = 70 msec) • In flight changes to 292 packets • Time to increase back to 583 is 20.4 seconds
npad results - 1 Data rate test: Pass! Pass data rate check: maximum data rate was 8.969226 Mb/s • Maximum data rate this link did achieve
Npad results - 2 Loss rate test: Pass! Pass: measured loss rate 0.035214% (2839 packets between loss events). • Loss caused by TCP overdriving the path FYI: To get 7 Mb/s with a 1460 byte MSS on a 22 ms path the total end-to-end loss budget is 0.282486% (354 packets between losses). • Worst case scenario for packet loss
Npad results - 3 Suggestions for alternate tests FYI: This path may even pass with a more strenuous application: Try rate=7 Mb/s, rtt=62 ms Try rate=8 Mb/s, rtt=48 ms Or if you can raise the MTU: Try rate=7 Mb/s, rtt=383 ms, mtu=9000 bytes Try rate=8 Mb/s, rtt=299 ms, mtu=9000 bytes • Helpful hints, you might do better or go farther at this rate
Npad results - 4 Network buffering test: Pass! Pass: The network bottleneck has sufficient buffering (queue space) in routers and switches. Measured queue size, Pkts: 36 Bytes: 52560 This corresponds to a 48.333600 ms drain time. To get 7 Mb/s with on a 22 ms path, you need 19250 bytes of buffer space. • Report on Network devices in path. A cheap switch would show up as something with small buffers • ‘[?]’ character indicates clickable icon for more details
Finding a Server • What? You don’t have one running at your site? • Install the Internet2 Network Performance Toolkit Knoppix Disk
Network Admin Tools • BWCTL – Bandwidth Control • Allows single person operation over wide area testing environment • Runs NLANR ‘iperf’ program • OWAMP – One way Delay Measurement • Advanced ‘ping’ command • Allows single person operation over wide area testing environment
BWCTL Highlights • You must pre-install BWCTL software package • New Internet2 default permits basic TCP test from any member • Sites can restrict access to ‘known’ remote users
Using BWCTL: commands bwctl -L90 -i2 -t20 -c bwctl.losa.net.internet2.edu bwctl -L90 -i2 -t20 -s bwctl.newy.net.internet2.edu • Bwctl = name of program • L90 = wait up to 90 seconds for a test • i2 = report intermediate results every 2 seconds • t20 – run test for 20 seconds • s name = remote end will send data to you • c name = you will send data to the remote host
3rd party testing: command bwctl -L90 -i2 -t20 -c bwctl.salt.net.internet2.edu -s bwctl.atla.net.internet2.edu • User can run a test between 2 remote hosts
Finding a Server • What? You don’t have one running at your site? • Install the Internet2 Network Performance Toolkit Knoppix Disk
OWAMP Highlights • You must pre-install OWAMP software package • User program is called ‘owping’ • New Internet2 default permits basic test from any member • Sites can restrict access to ‘known’ remote users
Using OWPING • owping owamp.salt.net.internet2.edu • owping = program name • owamp… = name of server • Output results • Separate statistics for both directions • Number of packets sent and lost • One-way delay statistics min/median/max • Number of IP hops in path • Number of packets that arrives out-of-order
Finding a Server • What? You don’t have one running at your site? • Install the Internet2 Network Performance Toolkit Knoppix Disk
Under Development • Emerging PerfSonar tool • Allows users to retrieve network path data from major national and international REN network
PerfSonar – Next Steps in Performance Monitoring • New Initiative involving multiple partners • ESnet (DOE labs) • GEANT (European Research and Education network) • Internet2 (Abilene and connectors)
PerfSONAR Services • Measurement Archive (MA) • Measurement Point (MP) • Lookup Service (LS) • Topology Service (TS) • Authentication Service (AS)
PerfSonar – Router stats on a path • Demo ESnet tool https://performance.es.net/cgi-bin/perfsonar-trace.cgi Paste output from Traceroute into the window and view the MRTG graphs for the routers in the path Author: Joe Metzger ESnet
Google it! • Enter “tuning tcp” into the google search engine. • Top 2 hits are: http://www.psc.edu/networking/perf_tune.html http://www-didc.lbl.gov/TCP-tuning/TCP-tuning.html