Active Measurement Boxes HADES (formerly called IPPM boxes) RIPE TTM boxes QoSMetrics boxes FrancoisXavierAndreu SimonLeinen 06 Jun 2005 06 ...

Active Measurement Tools Active measurement injects traffic into a network to measure properties about that network. ping a simple RTT and loss measurement ...

Active Queue Management (AQM) Packet switching nodes such as routers usually need to accomodate queues to buffer packets, when incoming traffic exceeds the available ...

Network Adapter and Driver Issues One aspect that causes many performance problems is adapater and NIC compatability issues (full vs half duplex as an example. The ...

DS3.3.3b Advanced Guide to Network Performance SimonLeinen 09 Apr 2006

Scaling Apache 2.x beyond 20,000 concurrent downloads ftp.heanet.ie is HEAnet's National Mirror Server for Ireland. Currently mirroring over 50,000 projects, it ...

Performance Hints for Application Developers Caveat Premature optimization is the root of all evil in programming E. Dijkstra... or was it C.A.R. Hoare... or was ...

Application Protocols TobyRodwell 28 Feb 2005

Host and Application Measurement Tools Network performance is only one part of a distrbuted system's perfomance. An equally important part is the performance od the ...

Automatically Generated Table of Contents Generated from UnrolledContents Apparently the %TOC% variable no longer expands include files in the topic, so this trick ...

Bandwidth Delay Product (BDP) The BDP of a path is the product of the (bottleneck) bandwidth and the delay of the path. Its dimension is "information", because bandwidth ...

Scripts for Bandwidth Measurement The information available from active tests such as BWCTL can be enhanced by embedding them in scripts which collect TCP stats before ...

Bandwidth Measurement Tools (Click on blue headings to link through to detailed tool descriptions and examples) Iperf Iperf is a tool to measure maximum TCP bandwidth ...

DS3.3.3a Basic Guide to Network Performance SimonLeinen 09 Apr 2006

Abstract This paper is a guide to network performance issues. We present the fundamentals of network performance, and then look at ways of investigating and remedying ...

Current Good Practice for Campus Networks ("Best Practice Guide") DRAFT Deliverable DS3.3.2, Part 2 Description from the GN2 Description of Work Deliverable description ...

Draft Outline for "Best Practice Guide" DS3.3.2 Part 2, Current Good Practice for Campus Networks Editor: RobertStoy "Upstream" connections Bandwidth and utilisation ...

BIC (Binary Increase Congestion Control) This was integrated into the Linux kernel in release 2.6.7, although it was ineffective until 2.6.11.7 due to an implementation ...

BitTorrent BitTorrent is an example of a peer to peer file sharing protocol. It employs local control mechanisms to optimize the global problem of replicating a ...

LAN Broadcast Domains While switches help network performance by reducing collision domains, they will permit broadcasts to all users and multicasts to specific groups ...

OS Specific Configuration Hints: BSD Variants Performance tuning By default, earlier BSD versions use very modest buffer sizes and don't enable Window Scaling by ...

Simple Script for Restarting bwctld #!/bin/bash #This script stops and restarts bwctld bwctld pid `cat /var/run/bwctld.pid` echo "Killing $bwctld pid ..." sudo kill ...

#!/bin/bash # This script runs a bwctl test and, concurrently runs the 'ss' (show socket command) # at regular intervals during the test. It is recommended the 'ss ...

Example Result of BWCTL/SS Script The command './testtcptest.sh mp1.gen.ch.geant2.net 20 0.2' was run on mp1.fra.de.geant2.net, a GÉANT2 Measurement Point device, ...

Changes required because ss output now spans two lines #!/bin/bash # This script runs a bwctl test and, concurrently runs the 'ss' ("show socket" command), # ...

BWCTL BWCTL is a command line client application and a scheduling and policy daemon that wraps throughput measurement tools such as iperf , thrulay , and nuttcp ...

BWCTL and BWCTLD For full information on these tools, see the 'Man' pages, either in the applications or on line at http://e2epi.internet2.edu/bwctl/manpages.html ...

Default bwctld.limits File # # $Id: BwctlToolInfoLimFile.txt,v 1.1 2006/09/19 13:01:00 TobyRodwell Exp www data $ # ######################################### ...

Bwctld.limits Parameters See default bwctld.limits file for examples of use allow open mode This limit is only useful if the class is assigned to a netmask. It ...

CARNET (HR) Public Tools Bandwidth meter Visual PerfSONAR SimonLeinen 24 Nov 2006

CSMA/CD CSMA/CD stands for Carrier Sense Multiple Access/Collision Detection . It is a central part of the original Ethernet design and describes how multiple stations ...

"Chatty" Protocols A common problem with naively designed application protocols is that they are too "chatty", i.e. they imply too many "round trip" cycles where one ...

Checksum Offload A large part of the processing costs related to TCP is the generation and verification of the TCP checksum. Many Gigabit Ethernet chipsets include ...

OS specific Tuning: Cisco IOS Cisco IOS specific Tuning TCP MSS Adjustment The TCP MSS Adjustment feature enables the configuration of the maximum segment ...

LAN Collisions In some legacy networks, workstations or other devices may still be connected as into a LAN segment using hubs. All incoming and outgoing traffic is ...

Compound TCP Compound TCP was developed by Microsoft for the "Vista" operating system. It combines two types of congestion control mechanisms: TCP Reno's loss based ...

Content Delivery Networks (CDNs) Content Delivery Networks are operated to support the distribution of content from content owners to content consumers. A typical ...

Content Distribution Networks (CDNs) See ContentDeliveryNetworks. SimonLeinen 19 Apr 2005 20 Nov 2008

Cricket Cricket is a flexible tool developed to visualise and understand network traffic. It uses RRDTool for data storage and can monitor any host or network based ...

CUBIC An enhancement of BIC, this was added to the BIC implementation in Linux 2.6.16, and became the default algorithm in 2.6.19. In Linux 2.6.26, the algorithm ...

DS3.3.2 PERT User Guide and Best Practice Guide DS3.3.2 was submitted to the EU at the end of August 2005, and is available from the G Eacute;ANT2 Document Library ...

DTrace DTrace is a dynamic tracing framework developed by Sun for their Solaris and OpenSolaris operating systems. It allows tracing many aspects of execution across ...

Using DAG cards to capture packets Two utilities (dagsnap and dagconvert) can be used to capture packets with a DAG card and convert captured data to tcpdump format ...

Delay Variation ("Jitter") Delay variation or "jitter" is a metric that describes the level of disturbance of packet arrival times with respect to an "ideal" pattern ...

Differentiated Services (DiffServ) From the IETF's Differentiated Services (DiffServ) Work Group's site: The differentiated services approach to providing quality ...

Digital Video Transport System DVTS uses RTP over UDP to send high quality video and sound over an IP network. It has a quality similar to MPEG 2 but is cheaper and ...

DFS (Distributed File System) and AFS (Andrew File System) AFS was developed in the 1980s at Carnegie Mellon University (CMU). It had several significant differences ...

dummynet A flexible tool for testing networking protocols, dummynet has been included in FreeBSD since 1998. A description of dummynet can be found on its homepage ...

Duplex Modes and Auto Negotiation A point to point Ethernet segment (typically between a switch and an end node, or between two directly connected end nodes) can operate ...

ETrace ETrace is a traceroute variant with customizable probe packets. It supports various protocols (ICMP, TCP, UDP and other IP protocols). A lot of the IP/TCP/UDP ...

End System Performance In Research and Education networking, performance demanding applications are normally run on hosts which are located on or very near national ...

End System Hardware Issues Sometimes an end system's hardware is not sufficient to support the demands place on it by high speed networking. For example, a system ...

Linux Memory Usage The first few lines of the top shell command will show memory stats for a Linux system. admin@pm1 pert $ top top 18:57:48 up 134 days, 2:26 ...

Interrupt Processing Introduction When an end system NIC receives a packet it will raise an interrupt. This might be done immediately, or it might be delayed, depending ...

Considerations for Sizing TCP Buffers on Hosts Until very recently, it was the case that most operating systems shipped with default settings for TCP that were suitable ...

Traffic Shaping TCP by its nature creates bursty traffic. Traffic bursts are susceptible to packet loss as there is greater risk of them overfilling a network node ...

End System (Host) Tuning This section contains hints for tuning end systems (hosts) for maximum performance. Considerations for sizing TCP buffers Tuning ...

Ethernet Ethernet is now widely prevalent as a link layer technology for local area/campus network, and is making inroads in other market segments as well, for example ...

Ethernet Flow Control The purpose of Ethernet Flow Control (which is defined in IEE 802.3x) is to allow an Ethernet receiving station to tell a sending station to ...

An evil middlebox is a transparent device that sits inbetween an end to end connection that disturbs the normal end to end traffic in some way. As you can not see ...

Explicit Congestion Notification (ECN) The Internet's end to end rate control schemes (notably the TransmissionControlProtocol (TCP)) traditionally have to rely on ...

Explicit Congestion Protocol (XCP) Explicit Congestion Protocol (XCP) is a new congestion control protocol developed by Dina Katabi from MIT Computer Science Artificial ...

Explicit Control Protocol (XCP) The eXplicit Control Protocol (XCP) is a new congestion control protocol developed by Dina Katabi from MIT Computer Science Artificial ...

TobyRodwell 28 Feb 2005

FAST FAST is an enhanced congestion control algorithm for TCP, which utilizes delay variations as congestion indications, not just packet loss. It was developed ...

Public Tools Looking Glass (IPv4/ IPv6) allows to run ping, traceroute and bgp commands in GIGAPIX router Multicast Beacon (IPv6) IPv6 multicast reachability ...

File Transfer A common problem for many scientific applications is the replication of often large data sets (files) from one system to another, or to several others ...

File2Net/Net2File Parts of the eVLBI community use file transfer utilities from the Mark5 software suite distributed by MIT Haystack Observatory. The simplest of ...

TCP Flow Control Note: This topic describes the Reno enhancement of classical "Van Jacobson" or Tahoe congestion control. There have been many suggestions for improving ...

fping fping is a ping like program which uses the Internet Control Message Protocol (ICMP) echo request to determine if a host is up. fping is different from ...

G Eacute;ANT Tools PERT Staff can access a range of tools for checking the status of the G Eacute;ANT network by logging into http://stats.geant.net/ (if a member ...

Using GÉANT2 bwctl Servers NRENs may request access to BWCTL servers in order to run iperf tests to and from their own BWCTL servers (and even to run tests between ...

Example section bwcltd.limits File The text below shows the configuration of a 'troubleshooter' group, and then assigning the GÉANT2 bwctl server (which are all within ...

Ports used (and allowed) by GÉANT2 bwctl servers 4823 , 5001 5010, 33400 33499, 50004 50014 , 32000 32999 (Ports in bold are those used by GÉANT2) TobyRodwell ...

Troubleshooting Procedures PERT Troubleshooting Procedures, as used in the earlier centralised PERT in GN2 (2004 2008), are laid down in GN2 Deliverable DS3.5.2 (see ...

Glossary of Abbreviations and Acronyms ACK TCP acknowledgement packet AIMD Additive Increase/Multiplicative Decrease AQM Active Queue Management ASIC Application ...

GridFTP GridFTP is a file transfer protocol that uses parallel streams to provide good bulk performance over LFNs. The underlying transport can be TCP or UDT. GridFTP ...

GRNET Network Monitoring Tools Network Monitoring Tools Portal includes links to: Looking Glass (IPv4 and IPv6 ping, trace, bgp) MRTGs Network Graphs ...

General Background and Context Motivation and Purpose The GN2 Performance Enhancement and Response Team (PERT) is a virtual team committed to helping academic users ...

HADES HADES (HADES Active Delay Evaluation System) is the new name for DFN's IPPM system TobyRodwell 25 Jan 2007

HADES Hades Active Delay Evaluation System (HADES) devices (previously called IPPM devices) were developed by the WiN Labor at RRZE (Regional Computing Centre Erlangen ...

Hamilton TCP Hamilton TCP (H TCP), developed by Hamilton Institute, is an enhancement to existing TCP congestion control protocol, with good mathematic and empirical ...

Head of line blocking An EthernetSwitch connects a number N of input ports to a number M of output ports (usually N M ) by some sort of SwitchingFabric. ...

Public Tools MRTG Traffic statistics for core and access circuits Map Weathermap based on MRTG Looking Glass IPv4 and IPv6 Private Tools Contact ...

HighSpeed TCP HighSpeed TCP is a modification to TCP congestion control mechanism, proposed by Sally Floyd from ICIR (The ICSI Center for Internet Research). The main ...

Bus architecture and performance Network (and other) interfaces attach to the motherboard using some kind of interconnect. Typically a motherboard has a couple of ...

Symptom : Accessing a specific web site which contains javascript, is very slow (around 30 seconds for one page) Analysis Summary : HTTP traffic is split between ...

Public Tools Looking Glass (IPv4) allows to run ping, traceroute and bgp commands MTR ping traceroute Traffic statistics for core and access circuits ...

IP Performance Metrics (IPPM) Working Group An IETF Working Group that has been chartered in 1997 to define metrics for network performance. Over the years, the group ...

Integrated Services (IntServ) IntServ was an attempt by the IETF to add Quality of Service differentiation to the Internet architecture. Components of the Integrated ...

Interactive Applications The user perceived performance of interactive applications is often determined by very different parameters than for "batch" applications ...

Interface queue lengths txqueuelen The txqueuelen parameter of an interface in the Linux kernel. It limits the number of packets in the transmission queue in ...

Internet2 Land Speed Record The Internet2 Land Speed Record is an open and ongoing competition that should foster pushing the performance boundaries of high speed ...

Internet Engineering Task Force (IETF) The IETF is an open community setting standards for the Internet community. Various Working Groups (WGs) use processes around ...

Interrupt Coalescence (also called Interrupt Moderation , Interrupt Blanking , or Interrupt Throttling ) A common bottleneck for high speed data transfers is the ...

TOS (type of service) TOS byte 7 6 5 4 3 2 1 0 8 bits IP precedence PREC upper 3 bits DiffServ code point DSCP ...

Iperf Iperf is a tool to measure TCP throughput and available bandwidth, allowing the tuning of various parameters and UDP characteristics. Iperf reports bandwidth ...

Iperf start stop script Usage: iperfd , or iperfd 'query' shows all iperf servers which have their PIDs saved (and therefore can be stopped easily) whilst 'query ...

Iperf Example As you might expect, running iperf in UDP mode only generates traffic in one direction (other than the server report packet, which is a single UDP packet ...

ISTF Network Monitoring Tools Network Monitoring Tools Portal includes links to: Looking Glass (IPv4 and IPv6 ping, trace, bgp) Cacti Network Graphs ...

Public Tools Looking Glass (IPv4) allows to run ping, traceroute and bgp commands Many local IUCC stats Interfaces, upstreams, environment, Usenet news ...

jnettop Captures traffic coming across the host it is running on and displays streams sorted by the bandwidth they use. The result is a nice listing of communication ...

A Jitter Buffer is a mechanism employed by real time media applications to eliminate the effects of DelayVariation introduced by networks and end systems. Packets ...

Support for Large Frames/Packets ("Jumbo MTU") On current research networks (and most other parts of the Internet), end nodes (hosts) are restricted to a 1500 byte ...

"Kibibyte" (KiB) A Kibibyte corresponds to 210 1024 bytes. SimonLeinen 29 Sep 2007

Large Receive Offload (LRO) Large receive offload is a performance optimization between the networking stack and device drivers. With LRO, the networking code in ...

Large Send Offload (LSO) This feature is also known as "Segmentation Offload", "TCP Segmentation Offload (TSO)", " TCP Multidata Transmit (MDT), or "TCP Large Send ...

Large TCP Windows In order to achieve high data rates with TCP over "long fat networks", i.e. network paths with a large bandwidth delay product, TCP sinks (that is ...

Latest News Efforts to increase TCP's initial congestion window The 78th IETF meeting is taking place in Maastricht this week, and one of the many topics being discussed ...

LFT (Layer Four Traceroute) lft is a variant of traceroute that uses per default TCP and port 80 to go through packet filter based firewalls. Example 142:/home ...

LBE (Less than Best Effort) Service Less Than Best Effort (LBE) is a new service available on GEANT/GN2 and some of the NRENs connected to it. It uses DiffServ with ...

libtrace libtrace is a library for trace processing. It supports multiple input methods, including device capture, raw and gz compressed trace, and sockets; and mulitple ...

Linux Specific Network Performance Tuning Hints Linux has its own implementation of the TCP/IP Stack. With recent kernel versions, the TCP/IP implementation contains ...

Linux Traffic Shaping Example The following is a sample configuration from a real network. The host in question is connected via 100 Mb/s Fast Ethernet to a "branch ...

These are living documents The User Guide and Best Practice Guide are based on material from the PERT Knowledge Base . The PERT Knowledge Base is expected to ...

"Long Fat Networks" (LFNs) Long Fat Networks (LFNs, pronounced like "elephants") are networks with a high bandwidth delay product. One of the issues with this type ...

LoriotPro Trace Route Discover Plugin LoriotPro is a network management framework commercially distributed by Luteus, a French company. It runs on various versions ...

Lustre File System This is a high performance distributed file system for HPC clusters. Server components include Metadata Servers (MDS) and Object Storage Servers ...

OS Specific Configuration Hints: Mac OS X As Mac OS X is mainly a BSD derivative, you can use similar mechanisms to tune the TCP stack see under BsdOSSpecific. ...

Management Information Base (MIB) This term describes the set of management information accessible through a network management protocol such as SNMP. SNMP uses a ...

MTR (Matt's TraceRoute) mtr (Matt's TraceRoute) combines the functionality of the traceroute and ping programs in a single network diagnostic tool. Example ...

Maximum Transmission Unit (MTU) The MTU (or to be exact, 'protocol MTU') of a link (logical IP subnet) describes the maximum size of an IP packet that can be transferred ...

Maximum Transmission Unit (MTU) The MTU (or to be exact, 'protocol MTU') of a link (logical IP subnet) describes the maximum size of an IP packet that can be transferred ...

Measurement Tools Traceroute like Tools: traceroute, MTR, PingPlotter, lft, tracepath, traceproto There is a large and growing number of path measurement tools derived ...

Terminology A middlebox is a device that has the ability to actively or passively participate in a communication in which it does not act as an end system. Typically ...

Mobile IP In wired and wireless environments where users carry mobile devices across multiple LAN subnets those devices often change their IP addresses, because IP ...

Modern Linux Traceroute This GPL version of traceroute by Dmitry Butskoy includes everything you could wish for in a traceroute or layer 4 traceroute. It requires ...

MRTG (Multi Router Traffic Grapher) The Multi Router Traffic Grapher (MRTG) is a tool designed to monitor the traffic load on network links. In practice, it can be ...

Network Adapters Demultiplexing to Multiple Threads Over the past years, multi threaded and multi core machines have become dominant, not just in the high performance ...

Multicast Traceroute ( mtrace ) The original traceroute idea of sending probe packets with increasing TTLs and evaluating the generated ICMP "destination unreachable ...

NREN Tools and Statistics A list of all NRENs and other networks who offer statistics/info for their networks and/or have tools available for PERT staff to use. ...

NANOG Traceroute Ehud Gavron maintains this version of traceroute. It is derived from the original traceroute program, but adds a few features such as AS (Autonomous ...

ICSI Netalyzr Netalyzr is a java applet developed by ICSI which will let you do a whole bunch of Internet connectivity tests. This can be used to determine the Internet ...

Netdude Netdude (Network Dump data Displayer and Editor) is a framework for inspection, analysis and manipulation of tcpdump (libpcap) trace files. References ...

Linux netem (introduced in recent 2.6 kernels) NISTnet works only on 2.4.x Linux kernels, so for having the best support for recent GE cards, I would recommend kernel ...

NetLogger NetLogger (copyright Lawrence Berkeley National Laboratory) is both a set of tools and a methodology for analysing the performance of a distributed system ...

Netflow based Tools Netflow: Analysis tools to characterize the traffic in the network, detect routing problems, security problems (Denials of Service), etc. There ...

Netperf A client/server network performance benchmark (debian package : netperf). Example coming soon References Home page: http://www.netperf.org/netperf ...

Sizing of Network Buffers Where temporary congestion cannot be avoided, some buffering in network nodes is required (in routers and other packet forwarding devices ...

NDT (Network Diagnostic Tool) NDT can be used to check the TCP configuration of any host that can run Java applets. The client connects to a Web page containing a ...

Network Emulation Tools In general, it is difficult to assess performance of distributed applications and protocols before they are deployed on the real network, because ...

Network File System (NFS) NFS was developed in 1984 by Sun, who bundled it with their SunOS operating system. It is layered on top of Sun RPC (Remote Procedure Call ...

Windows Network Monitor Microsoft Research has developed its own Wireshark variant called the Network Monitor. It's current version at time of writing is V3.3 which ...

Network Performance Metrics There are many metrics that are commonly used to characterize the performance of networks and parts of networks. We present the most important ...

Network Protocols This section contains information about a few common Internet protocols, with a focus on performance questions. Transmission Control Protocol ...

Network proxies or so called "WAN accelerators" have sometimes been deployed especially in the commercial world to speed up communications between remote sites. They ...

Network Tuning This section describes a few ways that can be used to improve performance of a network. Active Queue Management (AQM) is a class of router mechanisms ...

Larry Dunn from Cisco has set up a portable network emulator based on NISTnet. He wrote this on it: My systems started as 2x100BT, but not for the (cost) reasons ...

News Archive (2006) For current news, check out the LatestNews section New TWiki software I'm in the process of upgrading the TWiki software to the new 4.0.5 release ...

News Archive (2007) For current news, check out the LatestNews section SC'07 Bandwidth Challenge results According to this press release from Indiana University, ...

News Archive (2008) For current news, check out the LatestNews section Recordings and other Materials from eduPERT Training Event now available In late November 2008 ...

One way Delay (OWD) One way delay is the time it takes for a packet to reach its destination. It is considered a property of network links or paths. RFC 2679 contains ...

Calculating required TCP buffer sizes Regardless of operating system, TCP buffers must be appropriately dimensioned if they are to support long distance, high data ...

Operating System Specific Configuration Hints This topic points to configuration hints for specific operating systems. Microsoft Windows Linux BSD ...

TobyRodwell 24 Jan 2005 The following information is courtesy of Jeff Boote, from Internet2, and was forwarded to me by Nicolas Simar. Operating Systems (OSs) have ...

OWAMP (One Way Active Measurement Tool) OWAMP is a command line client application and a policy daemon used to determine one way latencies between hosts. It is an ...

PPP over Ethernet (PPPoE) PPP over Ethernet is a popular encapsulation method for IP packets in the ADSL world. It can be used when a computer or router/NAT is connected ...

Packet Loss Packet loss is the probability of a packet to be lost in transit from a source to a destination. A One way Packet Loss Metric for IPPM is defined in ...

Packet Reordering The Internet Protocol (IP) does not guarantee that packets are delivered in the order in which they were sent. This was a deliberate design choice ...

Packet Capture and Analysis Tools: Detect protocol problems via the analysis of packets, trouble shooting tcpdump: Packet capture tool based on the libpcap ...

Packetization Delay Packetization is an important contributor to delay, in particular in packet voice applications such as Voice over IP. Thanks to advances in ...

Paris Traceroute "Paris traceroute" is a traceroute variant that detects different types of load balancing approaches in the network, and tries to enumerate parallel ...

Passive Measurement Tools Network Usage: SNMP based tools retrieve network information such as state and utilization of links, router CPU loads, etc.: MRTG, ...

Path Analyzer Pro Commercial system integrating traceroute with graphical display, DNS and whois lookups. Works on both Windows and MacOS X. Free trial license available ...

Path MTU The Path MTU is the Maximum Transmission Unit (MTU) supported by a network path. It is the minimum of the MTUs of the links (segments) that make up the path ...

pathping Seems to be similar to mtr, but for Windows systems. This is included in at least some modern versions of Windows. References "pathping", Windows XP ...

pchar Characterize the bandwidth, latency and loss on network links. (See below the example for information on how pchar works) (debian package : pchar) Example ...

Notes About Performance This part of the PERT Knowledgebase contains general observations and hints about the performance of networked applications. User Perceived ...

Performance Case Studies Detailed Case Studies This section describes some case studies in which alterations to end systems, applications or network elements improved ...

Performance friendly I/O interfaces read()/write() vs. mmap()/write() vs. sendfile() For applications with high input/output performance requirements (including ...

Network Performance People There are a many individuals who contributed to the field of network performance. A few of them are listed here if you think of someone ...

PERT Case Histories Publicly available information on each case the PERT investigates is given in the pages listed below Case # Title Client Opened ...

NERSC Case Study: Stateful Firewall Background This case study was presented at Joint Techs 2006 (Albuquerque, NM) by Brian Draney of NERSC. NERSC is the US DoE ...

NERSC Case Study: Stateful Firewall Background This case study was presented at Joint Techs 2006 (Albuquerque, NM) by Brian Draney of NERSC. NERSC is the US DoE ...

Tuneable Linux Kernel IP Parameters See the TCP Man Pages and Oscar Andreasson's Ipsysctl Tutorial for more information on Linux Kernel parameters. Changes in newer ...

PERT Training Workshop, 19 21 September 2007 On 19 21 September 2007, a training workshop on Setting up and Operating a PERT took place at the SWITCH offices in ...

eduPERT Training, Zurich, 27/28 November 2008 An eduPERT Training workshop was held in the SWITCH Offices in Zurich, Switzerland on 27 28 November 2008. Slides ...

PingPlotter PingPlotter by Nessoft LLC combines traceroute, ping and whois to collect data for the Windows platform. Example A nice example of the tool can be found ...

Difference in ping size between Juniper and Cisco routers Juniper Let's take a POS interface on Junipers: nocview@RT1 BO1 RE0 show interfaces so 1/3/0.0 Logical ...

ping Ping sends ICMP echo request messages to a remote host and waits for ICMP echo replies to determine the latency between those hosts. The output shows the Round ...

TCP Limitation Models Mathis Equation The Mathis equation states that, for a small loss rate (less than 1%), the maximum achievable throughput of a TCP connection ...

Premium IP Premium IP is a new service available on GEANT/GN2 and some of the NRENs connected to it. It uses DiffServ, and in particular the EF (Expedited Forwarding ...

Propagation Delay The propagation delay is the time it takes for a signal to propagate. It depends on the distance traveled, and the specific propagation speed ...

PSNC Network Monitoring Tools Public Tools Weathermap PIONIER Weather Map Network tools Looking Glass and PChar Private Tools Contact PSNC NOC for ...

Opera Oberta: preliminary measurements PTS Case 10 Background Ecole Polytechnique Fédérale de Lausanne (EPFL) wanted to receive high quality multicast transmissions ...

Large packets lost between Onsala and Stockholm PTS Case 11 Background Onsala observatory contacted the PERT about a problem with traceroute it did not work ...

Peering problem with AS3300? PTS Case 12 Synopsis IUCC reported large packet loss on the path between IL and AS3300 (Eqip, France), which is via GN2. Traceroute ...

Low performance between Budapest and Trieste PTS Case 13 Synopsis HUNGARNET reported poor performance between Budapest and the International Centre for Theoretical ...

Below expected throughput, Budapest to New York PTS Case 14 Synopsis DANTE contacted the PERT to report they were experiencing below expected TCP throughput between ...

ITER VPN Connection PTS Case 15 Synopsis The ITER project reported performance problems between their sites in Garching and Cadarache. Their connection ran over ...

Routing to USA going via Japan PTS Case 16 Synopsis IUCC reported a sub optimal path to a network in the US, that was being routed via GÉANT2 then TEIN2. It was ...

Slow file transfer UK to various European DEISA sites PTS Case 17 Synopsis The European Centre for Medium range Weather Forecasting (ECMWF), which is based in Reading ...

Slow file transfers between TIGO (CL) and JIVE (NL) PTS Case 18 Background JIVE was involved in a program to track the SMART 1 spacecraft in its orbit around the ...

Poor throughput from London to New York GN2 MP PTS Case 19 Background TCP throughput below expected value has been experienced between GN2 Measurement Point in ...

Low data throughput for e VLBI project PTS Case 2 Background Very Long Baseline Interferometry (VLBI) is a technique in radio astronomy for correlating data from ...

Problem in aggregating data streams Background As part of radio astronomy applications testing, 512 Mbps real time data were transmitted (streamed) from Jodrell Bank ...

Poor TCP performance on a L2 circuit PTS Case 21 Synopsis Poor TCP performance was observed on a L2 circuit between In2p3 and Fermilab. After tuning their end ...

TIGO Europe throughput check post REUNA upgrade PTS Case 22 Synopsis Once the capacity of the link between University de Concepcion and REUNA was upgraded from ...

Congested Links to SHERN PTS Case 23 Synopsis The EUMEDCONNECT NOC reported suspicious behaviour seen on the 4 E1 links to SHERN (the Syrian NREN) the RED (Random ...

Thruput: GSFC to TAU PTS Case 24 Synopsis GSFC is NASA's Goddard Space Flight Center, in Maryland, near Washington DC. ENPL is the EOS (Earth Observation System ...

Slow downloads from GEANT2 webserver to SWITCH PTS Case 25 Synopsis After downloading all the monthly reports from the GN1 project, a SWITCH user appeared to get ...

Italy to Japan space agencies low throughput PTS Case 26 Synopsis The European Space Agency in Italy ran data transfer tests between its site near Rome (ESA ESRIN ...

Bandwidth problem at Hephy Vienna PTS Case 27 Synopsis During operations with the GRIDFTP application, only 2.2Mbps were acheived between CERN and Austria, and ...

Low level packet loss on GÉANT2 DK DE backbone circuit PTS Case 29 Synopsis Whilst investigating an unrelated router issue (GEANT2 TT #1047863), low level packet ...

DEISA TCP throughput suffers from cross traffic PTS Case 3 Synopsis DEISA is a consortium of leading national supercomputing centres that currently deploys and ...

ESA Performance problem PTS Case 30 Synopsis European Space Agency started to have performance problems doing transfers between Spain (Maspalomas Satellite Gorund ...

Throughput between JIVE and Shanghai Observatory PTS Case 31 Synopsis Project ORIENT has leased an STM 16 circuit between the GEANT2 DK !PoP and the TEIN2 BJ !PoP ...

TCP throughput Problem on 150ms path between 10GE connected endsystems PERT Case 32 Synopsis Users at the Technical University in Dresden want to achieve Multi ...

Throughput tests (IT IN) for APAN Medical Working Group PTS Case 33 Synopsis The APAN Medical Working Group use DVTS for remote medical activities. Their experience ...

Packet loss for packets sent over GEANT2 Telia PTS Case 34 Synopsis RESTENA reported noticeable traffic loss in their Internet service. Independent of their ...

A.3 Performance Problems between FNAL and DESY PTS Case 4 Background The German Electron Synchrotron (DESY) is one of the leading accelerator centres in the world ...

Low throughput MU LSU PTS Case 5 Synopsis The PERT was contacted by CESNET who reported low IP throughput (about 32Mbps at UDP) between workstation at Masaryk ...

Slow upload from DANTE offices to PERT TWiki at SWITCH PTS Case 6 Synopsis DANTE staff noted that uploading documents to a SWITCH server was taking an unacceptably ...

Poor ftp performance to ftp.ncbi.nlm.nih.gov PTS Case 7 Synopsis IUCC users reported they were only able to get about 3Mbps FTP throughput to NIH (US National Institutes ...

DEISA Teragrid performance PTS Case 8 Synopsis DEISA contacted the PERT via DANTE to request help with the poor TCP performance they were experiencing in data transfers ...

DEISA decrease in throughput between IDRIS and other sites PTS Case 9 Synopsis Beginning in October, DEISA observed a decrease in the achievable throughput (as ...

Duplex Mismatch MDM Case 1 Background Observations Investigation Outcome TobyRodwell 26 Feb 2008

Throughput Limiitation MDM Case 2 Background Observations Investigation Outcome TobyRodwell 26 Feb 2008

Interface Errors MDM Case 3 Background Examples TobyRodwell 26 Feb 2008

QoSMetrics Boxes This measurement infrastructure was built as part of RENATER's M trologie efforts. The infrastructure performs continuous delay measurement between ...

Quick Start TCP Quick Start TCP was codified in an Experimental RFC (RFC 4782) in 2007. Based on explicit feedback from the network, it can use a large Initial Window ...

Random Early Detection (RED) RED is the first and most widely used Active Queue Management (AQM) mechanism. The basic idea is this: a route controlled by RED samples ...

Rate Control Protocol (RCP) Developed by Nandita Dukkipati and Nick McKeown in Stanford University, RCP aims to emulate processor sharing(PS) over a broad range of ...

Reading list for CMs List of relevant and useful PERT documents and URLs with short description PERT Trouble Ticket System PERT Trouble ticket system note that ...

Real Time Control Protocol RTCP provides out of band control information for an RTP flow. In particular, Receiver Reports include reception quality indications ...

Real Time Transport Protocol (RTP) RTP (RFC 3550) is a generic transport protocol for real time media streams such as audio or video. RTP is typically run over UDP ...

Related Efforts There are many other places where useful information about network performance can be found. Here is a small selection feel free to add more references ...

RENATER Network Monitoring Tools Public Tools Weathermap IPPM Measurements IPv4 IPv6 : delay, jitter, hops number and packets loss between few Renater ...

How to Report a Performance Problem As you investigate a network performance issue, whether as a network user, support person such as a systems or network administrator ...

RIPE Test Traffic Project (TTM) bla bla bla SimonLeinen 11 May 2005

RIPE TTM Boxes Summary The Test Traffic Measurements service has been offered by the RIPE NCC as a service since the year 2000. The system continuously records ...

Round Trip Time (RTT) Round trip time (RTT) is the total time for a packet sent by a node A to reach is destination B, and for a response sent back by B to reach A ...

Router Architectures Basic Functions Basic functions of an IP router can be divided into two main groups: The forwarding plane is responsible for packet forwarding ...

Router Hardware Limitations Active network devices such as routers have performance limitations along various dimensions which, when exceeded, cause performance degradations ...

Router Performance In order to evaluate the performance impact of active network devices such as routers, an understanding of the underlying platforms is required ...

RRDtool RRDtool is a data logging and graphing application. It is based on a Round Robin Database which stores and displays time series data. It can be used via simple ...

RUDE/CRUDE RUDE is a package of applications to generate and measure UDP traffic between two endpoints (hosts). The rude tool generates traffic to the network ...

Sally Floyd Sally co invented RandomEarlyDetection with VanJacobson, proposed ExplicitCongestionNotification, and works on the HS TCP (High Speed TCP) variant of the ...

Scalable TCP One of the advantages of Scalable TCP is that its recovery time (after loss) is only proportional to the RTT, and not also proportional to the sending ...

Scamper The Scamper tool was developed by Matthew Luckie of the WAND group at the University of Waikato (New Zealand) to facilitate large scale measurements over ...

Secure Shell (SSH) SSH is a widely used protocol for remote terminal access with secure authentication and data encryption. It is also used for file transfers, using ...

TCP Selective Acknowledgements (SACK) Selective Acknowledgements are a refinement of TCP's traditional "cumulative" acknowledgements. SACKs allow a receiver to acknowledge ...

Serialization Delay (or Transmission Delay ) Serialization delay is the time it takes for a unit of data, such as a packet, to be serialized for transmission on a ...

Scaling Servers While much of the focus of the early PERT has been on the performance of individual TCP connections, frequently one is interested in the aggregate ...

SIFTR Statistical Information for TCP Research Written by James Healy and Lawrence Stewart from Swinburne University in 2007, SIFTR is a kernel module for FreeBSD ...

SNMP The Simple Network Management Protocol was standardized by the IETF and is mostly used by operators to monitor various types of network equipment, as well as ...

SmokePing Smokeping is a software based measurement framework that uses various software modules (called probes) to measure round trip times, packet losses and availability ...

SmokeTrace Recent versions of the SmokePing active measurement and visualization tool include a new tool called SmokeTrace. Like SmokePing, SmokeTrace runs on a Web ...

SNMP based tools The Simple Network Management Protocol (SNMP) is widely used for device monitoring over IP, especially for monitoring infrastructure compoments of ...

SNMP Counter Wrap Many of the variables accessible through SNMP are counters . Counters in SNMP are never reset, and they don't have a defined starting value. So ...

Low Priority for SNMP Handling Devices such as routers often give low priority to the SNMP, which handles SNMP requests, compared to other tasks that are deemed more ...

Synagon SNMP Graphing Tool Overview Synagon is Python based program that collects and graphs firewall values from Juniper routers. The tool first requires the user ...

SNMP Update Lag When accessing management variables such as traffic counters over SNMP, you should be aware of the following: The SNMP agent on a managed device ...

Solaris snoop Sun's Solaris operating system includes a utility called snoop for taking and analyzing packet traces. While snoop is similar in spirit to tcpdump ...

OS Specific Configuration Hints: Solaris (Sun Microsystems) Planned Features Pluggable Congestion Control for TCP and SCTP A proposed OpenSolaris project foresees ...

Solaris traceroute (includes IPv6 functionality) On Sun's Solaris, IPv6 traceroute functionality is included in the normal traceroute program, so a separate "traceroute6 ...

Source Quench Source Quench is an ICMP based mechanism used by network devices to inform data sender that the packets can not be forwarded due to buffers overload ...

Specific Network Technologies This section of the knowledge base treats some (sub)network technologies with their specific performance implications. Ethernet ...

SCTP (Stream Control Transmission Protocol) SCTP is a transport protocol operating analogously to TCP or UDP on top of a connectionless packet service such as IP. ...

Sun hxbt STREAMS Module From an announcement on the OpenSolaris Networking community site: Hxbt is a Stream module/driver which emulates a WAN environment. It captures ...

Performance Related Tools at SWITCH IPv4 Looking Glass allows running IPv4 ping, traceroute, and several BGP monitoring commands on external border routers ...

Sync TCP Sync TCP is based on Reno's congestion control, but in addition adapts its cwnd according to changes in one way delay. The name "Sync TCP" is actually a misnomer ...

TCP Acknowledgements In TCP's sliding window scheme, the receiver acknowledges the data it receives, so that the sender can advance the window and send new data ...

Automatic Tuning of TCP Buffers Note: This mechanism is sometimes referred to as "Dynamic Right Sizing" (DRS). The issues mentioned under "Large TCP Windows" are ...

TCP Traceroute TCPTraceroute is a traceroute implementation that uses TCP packets instead of UDP or ICMP packets to send its probes. TCPtraceroute can be used in ...

Windows TCP Analyzer The Microsoft Windows SDK also contains a tool that can read the so called ESTATS of an open TCP connection. When running, you get a listing of ...

Modifying BIC/CUBI initial ssthresh value (As of 17 May 07) By default, in BIC adn CUBIC ssthresh is set to 100, so that when cwnd reaches this value it stops growing ...

tcpdump One of the early diagnostic tools for TCP/IP that was written by Van Jacobson, Craig Leres, and Steven McCanne. Tcpdump can be used to capture and decode ...

TCP Fusion TCP Fusion is a combination of a loss based and a delay based congestion control protocol similar to TCP Adaptive Reno and Compound TCP. It combines characteristics ...

High Speed TCP Variants There have been numerous ideas for improving TCP over the years. Some of those ideas have been adopted by mainstream operations (after thorough ...

TCP's Initial Congestion Window (ICW or IW) The initial version of TCP Congestion Control specified that the congestion window should start at the maximum segment ...

"NewReno" TCP NewReno is an improvement on Reno TCP, which improves the performance of handling loss of multiple segments in a single round trip time in the absence ...

TCP Offload Engines (TOEs) The idea of a TOE is to put the TCP implementation onto the network adapter itself. This relieves the computer's CPUs of handling TCP segmentation ...

To improve the TCP performance many enhancements on TCP were done in the last years. E.g. implementation of improved TCP flow control algorithms (larger default buffer ...

TCP Reno TCP Reno builds upon TCP Tahoe's congestion control, adding the Fast Recovery mechanism described in RFC 2001. With this mechanism, isolated packet losses ...

TCP Tahoe Part of the "Tahoe" release of the BSD (Berkeley Unix) distribution in 1988, this was the first TCP to implement congestion control. It represented the ...

TobyRodwell 06 Apr 2005 Active open connection A TCP connection that a host has initiated i.e it sent the SYN Passive open connection A TCP connection ...

Tcptrace Tcptrace is a tool to analyze TCP (and now, also UDP) sessions captured with TcpDump. It provides a lot of different kinds of statistics and information ...

TCP Vegas TCP Vegas uses variations in measured throughput as an indication of congestion, in addition to packet loss. It also uses finer timer granularity compared ...

tcpdump (examples) Here are some more tcpdump examples for some more advanced use cases. For simple usage examples, see the main tcpdump topic. Filter on protocol ...

EXPERIMENTAL 10Gbps TECHNOLOGY RESULTS AT FORSCHUNGSZENTRUM KARLSRUHE A world wide, distributed Grid Computing environment is currently under development for the upcoming ...

3D Traceroute 3D Traceroute performs repeated measurements of the RTTs along the path, and displays the results in a three dimensional graph in color. The program ...

Thrulay Thrulay is a measurement tool suggested by Stanislav Shalunov. It performs TCP throughput ("bandwidth") tests similar to many other tools, but at the same ...

TODO The PERT Knowledge Base is a Wiki, so it can be edited by anyone after a simple registration procedure. Here are a few suggestions for further developing the ...

tracefilter This tool is part of the ANA Spoofer Project, which tries to assess the extent to which "spoofing", i.e. using illegitimate source addresses, is possible ...

TraceFlow Traceflow is a proposed protocol to improve OAM (Operations and Management) of complex Layer 2/Layer 3 networks that may include LAG (Link Aggregation Groups ...

traceiface The traceiface tool was presented by John Kristoff at NANOG39. It uses an expanding ring search technique to find not only the ingress point of each ...

tracepath tracepath and tracepath6 trace the path to a network host, discovering MTU and asymmetry along this path. As described below, their applicability for ...

traceproto Another traceroute variant which allows different protocols and port to be used. "It currently supports tcp, udp, and icmp traces with the possibility ...

Traceroute Mesh Server Traceroute Mesh combines traceroutes from many sources at once and builds a large map of the interconnections from the Internet to the specific ...

Traceroute6 Traceroute6 uses the Hop Limit field of the IPv6 protocol to elicit an ICMPv6 Time Exceeded ICMPv6 message from each gateway ("hop") along the path to ...

Windows tracert All Windows versions that come with TCP/IP support include a tracert command, which is a simple traceroute client that uses ICMP probes. The (few ...

I found VanJacobson's announcement of the original traceroute (see TracerouteLikeTools) tool in the archives of the end2end interest mailing list for the year ...

traceroute like Tools Traceroute is used to determine the route a packet takes through the Internet to reach its destination; i.e. the sequence of gateways or "hops ...

Net::Traceroute::PurePerl A traceroute implementation in Pure Perl. Version 0.10 (August 2006, still current as of February 2007) only supports UDP or ICMP probes ...

Traceroute Servers There are numerous traceroute servers located throughout the world: http://www.traceroute.org/ http://www.bgp4.net/wiki/doku.php?id tools ...

TCP (Transmission Control Protocol) The Transmission Control Protocol (TCP) is the prevalent transport protocol used on the Internet today. It uses window based transmission ...

TTCP TTCP (Test TCP) is a utility for benchmarking UDP and TCP performance. Utility for Unix and Windows is available at http://www.pcausa.com/Utilities/pcattcp ...

UDT (UDP based Data Transfer Protocol) UDT is a UDP based protocol for bulk data transfers over high speed (1 Gb/s and above) wide area networks. It was used in the ...

Unrolled PERTKB Contents SimonLeinen 05 Feb 2006 by sex

UDP (User Datagram Protocol) The User Datagram Protocol (UDP) is a very simple layer over the host to host Internet Protocol (IP). It only adds 16 bit source and ...

Research Network User's Guide to Performance ("User Practice Guide") DRAFT Deliverable DS3.3.2, Part 1 Description from the GN2 Description of Work Deliverable ...

Draft Outline for "User Guide" DS3.3.2 Part 1, Research Network User's Guide to Performance Editor: AnnHarding Introduction Goals and target readership This is ...

Related Work The GN2 PACE PERT has produced another network performance guide, named Current Good Practice for Campus Networks or "Best Practice Guide". That document ...

User Perceived Performance User perceived performance of network applications is made up of a number of mainly qualitative metrics, some of which are in conflict with ...

Reliability Reliability is often the most important performance criterion for a user: The application must be available when the user requires it. Note that this ...

Responsiveness One of the most important user experiences in networking applications is the perception of responsiveness. If end users feel that an application is ...

Throughput/Capacity/"Bandwidth" Throughput per se is not directly perceived by the user, although a lack of throughput will increase waiting times and reduce the impression ...

Van Jacobson Traceroute, TCP, and DiffServ work Van Jacobson made vast contributions to networking, in particular related to performance. He wrote the original traceroute ...

Original Van Jacobson/Unix/LBL Traceroute The well known traceroute program has been written by Van Jacobson in December 1988. In the header comment of the source ...

VFER VFER is a tool for high performance data transfer developed at Internet2. Main goals for VFER include good out of the box performance user space ...

Videoconferencing TODO: talk about typical end system issues (narrowly striped shirts, lighting, room acoustics : ), delay introduced by (re)coding/compression in ...

VisualRoute Graphical tool for Traceroute and "ping" measurements, including IP Address Location and DNS Lookup functions. The program works under Windows. There ...

Telephony / Voice over IP (VoIP) Responsiveness/sensitivity to round trip time In its most common use for two party conversations, telephony is very sensitive to ...

Welcome to the home of PERTKB. This is a web based collaboration area for collecting the knowledge base of the eduPERT. If you're looking for topics to work on, check ...

See also the faster WebTopicList

// Web Home Changes Topics Index Search Welcome Register

This is a subscription service to be automatically notified by e mail when topics change in this PERTKB web. This is a convenient service, so you do not have to ...

Web100 Linux Kernel Extensions The Web100 project was established to close the " wizard gap" between what performance should be possible on modern research networks ...

PERTKB Web Preferences The following settings are web preferences of the PERTKB web. These preferences overwrite the site level preferences in ., and can be ...

TWiki's PERTKB web /view/PERTKB The PERTKB web of TWiki. TWiki is a Web Based Collaboration Platform for the Corporate World.

Statistics for PERTKB Web Month: Topic views: Topic saves: File uploads: Most popular topic views: Top contributors for topic save and ...

TCP Westwood An improvement on TCP Westwood, which adds a bandwidth estimation algorithm that "provides a measure of used bandwidth that is not aliased" (from an ...

TCP Westwood The current Standard TCP implementation rely on packet loss as an indicator of network congestion. The problem in TCP is that it does not possess the ...

Why Latency Is Important Traditionally, the metric of focus in networking has been bandwidth . As more and more parts of the Internet have their capacity upgraded ...

Window Based Transmission TCP is a sliding window protocol . The receiver tells the sender the available buffer space at the receiver (TCP header field "window" ...

TCP Window Scaling Option TCP as originally specified only allows for windows up to 65536 bytes, because the window related protocol fields all have a width of 16 ...

The window scaling problem Many firewalls check that a TCP segment falls within the expected window. If window scaling is in use but the firewall does not support ...

Windows Specific Host Tuning Next Generation TCP/IP Stack in Windows Vista / Windows Server 2008 / Windows 7 According to http://www.microsoft.com/technet/community ...

Wireshark Wireshark is a packet capture/analysis tool, similar to tcpdump but much more elaborate. It has a graphical user interface (GUI) which allows "drilling ...

Wireless LAN Wireless Local Area Networks according to IEEE 802.11 standards has become extremely widespread in recent years, in campus networks, for home networking ...

The "Wizard Gap" The Wizard Gap is an expression coined in the context of the Internet2's end to end performance initiative. It designates the difference between ...

Yeah TCP Heuristic TCP protocol gained from experimental evaluation of CUBIC, HSTCP, H TCP, Africa, Compound TCP. Algorithm Design Goals: Network capacity ...