This document describes achema for publishing network measurement data. It is assumed that the reader of this document is familiar with the GGF NM-WG document:A Hierarchy of Network Performance Characteristics for Grid Applications and Services, which defines a classification hierarchy for network measurements that are useful for Grid applications and services. Use of the schemas described in this document should facilitate the development of interoperable Grid services.
As an example of how such network measurements could be used in a Grid environment, we use the case of a Grid file transfer service. Assume that a Grid Scheduler determines that a copy of a given file needs to be copied to site A before a job can be run. Several copies of this file are registered in a Data Grid Replica Catalogue, so there is a choice of where to copy the file from. The Grid Scheduler needs to determine the optimal method to create this new file copy, and to estimate how long this file creation will take. To make this selection the scheduler must determine what is the best source (or sources) to copy the data from. Selecting the best source to copy the data from requires a prediction of future end-to-end path characteristics between the destination and each possible source. Accurate prediction of the performance obtainable from each source requires measurement of available bandwidth (both end-to-end and hop-by-hop), latency, loss, and other characteristics important to file transfer performance.
Report of a network measurement.
The hosts and mnetworks that are being tested.
What test was run, and how it was run, in order to collect the measurement.
Result of performing a network test
Wildcard result that allows any additional user content after the timestamp
Abstract base type for all result set types.
This kind of result set will allow an 'anyResult', which can in turn contain pretty much anything. Therefore, this allows results that are described in a different schema, or no schema at all, to be reported.
Result set for characteristics:path.delay.* . For more information on the details of these properties, see the following IETF documents: One way Delay (http://www.ietf.org/rfc/rfc2679.txt), Round Trip Delay (http://www.ietf.org/rfc/rfc2681.txt).
Delay on a path
Result set for indicated characteristic
Result set for indicated characteristic
Result set for characteristics: path.loss.roundTrip, path.loss.oneWay. For more information on the details of these properties, see the following IETF documents: One-way Loss: http://www.ietf.org/rfc/rfc2680.txt, Loss Patterns: http://www.ietf.org/rfc/rfc3357.txt
number of packets since the previous loss (See RFC3357)
number of groups of lost packets (See RFC3357)
percent of packets lost where if the distance between the lost packet and the previously lost packet is no greater than the "loss constraint" (See RFC3357)
total number of loss periods (See RFC3357)
number of packets in a burst of loss (See RFC3357)
number of packets between bursts of loss (See RFC3357)
number of packets lost out during the test
average packet loss (in percent) (See RFC2680)
Result set for indicated characteristic
Result set for indicated characteristic
For more information on the details of these properties, see the following IETF documents: Delay Variation (Jitter): http://www.ietf.org/rfc/rfc3393.txt
See RFC3393
See RFC3393
Result set for indicated characteristic
For more information on the details of these properties, see the following IETF documents: http://www.ietf.org/internet-drafts/draft-ietf-ippm-reordering-02.txt
See IPPM draft
number of positions out of order
result, in percent
Result set for indicated characteristic
Result set for indicated characteristic
which hop is the bottleneck ("tight link")
bandwidth of path
Result set for indicated characteristic
Result set for indicated characteristic
Result set for indicated characteristic
indication of what is the bottleneck (network, CPU, NIC, memory, disk, etc.)
Result set for indicated characteristic
Mean Time Between Failures
Number of periods when path was not available
Outage length
percent Availability = 100*(#unavailable cycles/total cycles)%
Result set for indicated characteristic
Number of hops
A (generic) statistic about a value. Head of the substitution group.
Count of measurements
Minimum measured value
Maximum measured value
Mean measured value
Median measured value
Standard deviation of measured values
The n-th percentile of the measured values (median is a special case of this, really)
The percentile to which the value applies
The+/- n%confidence interval (default n=95%) . For example, a value of "1.5" means +/- 1.5
Percent confidence for which the given value provides an interval.
Network measurement tool
Parameters that affect how a tool performs its measurement
Parameters for tests that run over IP, including TCP/IP and UDP over IP.
The duration of the test
ICMP or UDP or TCP
Size of packets sent on network
Number of packets sent on network
What algorithm is used to space packets (poisson or periodic)
Time between test packets, in seconds (for periodic tests)
IP version (this information is also available as an attribute of the IPAddress, do we need it both places?)
Type of service (IP precedence)
differentiated services code point
IP v6 option for QoS
The threshold used to distinguish between a large finite delay and loss
Amount of test traffic
Are bytes moving memory to memory or disk to disk?
Size of TCP buffers used
Reno, Vegas, HSTCP, ScalableTCP, etc
Number of parallel streams
Some details about a host system.
Description of a host operating system
A network interface on a host
This repeats the information already contained in the parent type NetworkPath. In the XML instance, those elements may be referenced here to avoid duplication.
The hardware which interfaces between the host and the network transmission medium
The logical connection of the card as seen by the Operating System. E.g. eth0 orhme0:1
Version and name of NIC
Version and name of the NIC firmware
Version and name of NIC driver
The protocol stack on this host. Since this is a stack, the first protocol is the highest layer, and the last one is the lowest layer. If this element is present, the address associated with this node is for the highest layer of the protocol stack.
Path between a network source and destination node
A network node
If the tool used at this node is in any way different from the one described in the methodology section, the differences are shown here.
Abstract network address
Internet Protocol address
Internet Protocol version (4 or 6)
Version for a piece of software, for example a measurement tool.
Name and version information for a piece of hardware
Description of a system clock
Standard name and type for the units of a value
Named model group for a double-precision floating point value
Statistic(s) associated with value(s)
Named model group for a string value
Named model group for a 32-bit signed integer value
Statistic(s) associated with value(s)
Named model group for a network hop value
Named model group for a network path value