Federation – Smokeping, PingER Integration
SLAC Contact: Dr. Roger Les Cottrell, Connie, Aziz
Domain and Scope
Latency is one of the most important factors in the networked environment. There are various tools and infrastructures available to measure the end-to-end Round Trip Time (RTT) using the ubiquitous Internet ICMP ping facility, e.g. PingER, NLANR/AMP and SmokePing. Typically these infrastructures make regular measurements, archive them, analyze (to extract various metrics such as RTT, loss, jitter etc.), visualize (e.g. using time series), and provide alarms when the performance significantly degrades.
Our long-term goal is to initiate the integration of the various infrastructures into a federation so the measurements can be shared, and different analysis and visualization techniques can be applied to yield more ubiquitous coverage and new insights. This will also reduce the need for sites to need to deploy multiple infrastructures to gain all the available benefits.
This project is concerned with the integration of SmokePing and PingER as part of building the federation of measurement infrastructures. Both the Smokeping and PingER infrastructures make regular measurements, maintain longterm data-stores, and draw time series graphs, giving information on the state of each network connection. PingER is mainly aimed at long term network performance (it has data going back 11 years) using about a dozen metrics, in particular for developing regions, seen from multiple sites worldwide with many aggregations of measurement/remote pairs being provided. SmokePing utilizes the RRDtool database, features a an eye-catching time series graphing facility, a smart alarm mechanism for network administrators, and is mainly aimed at looking at the near real-time performance measured by RTT and loss for multiple individual pairs.
The requirements for this project are to provide a useful level of interoperability between PingER and Smokeping. This is motivated by the needs and wishes of National Carriers who run both PingER and Smokeping and want to get the best of both worlds.
1. Study and evaluate SmokePing and PingER to understand how they work, and their pros and cons.
2. Work with the developers of SmokePing to understand how to modify it and extend the capability of SmokePing to derive parameters/metrics not currently supported such as ping unreachability, jitter, derived TCP throughput, conditional loss probability etc. Provide the ability to select and visualize these new metrics. Feed the modifications to SmokePing back into the development base.
3. Evaluate and provide the ability to feed SmokePing data to the Pinger analysis/visualization toolkit, so we can use the PingER aggregation techniques to characterize performance by regions, or affinity groups.
4. Modify PingER to enable it to utilize the SmokePing time series visualization tool.
Benefits and Outcomes
1. National carrier using one measurement infrastructure can easily deploy and use other measurement infrastructure without any difficulty.
2. PINGER community can use the data collected by SmokePing infrastructure to get bigger and better picture of the network.
3. SmokePing community can use the data collected by PINGER infrastructure to get bigger and better picture of the network.
4. Expertise with both most widely deployed network performance measurement infrastructures.