Effects of Interrupt Coalescence on Network Measurements Export

@incollection{citeulike:5739659, abstract = {Several high-bandwidth network interfaces use Interrupt Coalescence (IC), i.e., they generate a single interrupt for multiple packets sent or received in a short time interval. IC decreases the per-packet interrupt processing overhead. However, IC also introduces queueing delays and alters the ” dispersion” (i.e., interarrival time spacing) of packet pairs or trains. In this work, we first explain how IC works in two popular Gigabit Ethernet controllers. Then, we identify the negative effects of IC on active and passive network measurements. Specifically, we show that IC can affect active bandwidth estimation techniques, causing erroneous measurements. It can also alter the packet interarrivals in passive monitors that use commodity network interfaces. Based on the ” signature” that IC leaves on the dispersion and one-way delays of packet trains, we show how to detect IC and how to filter its effects from raw measurements. Finally, we show that IC can be detrimental to TCP self-clocking, causing bursty delivery of ACKs and subsequent bursty transmission of data segments.}, author = {Prasad, Ravi and Jain, Manish and Dovrolis, Constantinos}, citeulike-article-id = {5739659}, citeulike-linkout-0 = {http://www.springerlink.com/content/a6tyk34er2xr6bwm}, journal = {Passive and Active Network Measurement}, keywords = {\_2004, \_uanm, active\_measurement, ethernet, experimental, gigabit, interrupt, measurement, nic, owd, passive\_monitors, protocol, tcp, workaround}, pages = {247--256}, posted-at = {2009-09-08 12:17:39}, priority = {2}, title = {Effects of Interrupt Coalescence on Network Measurements}, url = {http://www.springerlink.com/content/a6tyk34er2xr6bwm}, year = {2004} }

TY - CHAP ID - citeulike:5739659 L3 - citeulike-article-id:5739659 N2 - Several high-bandwidth network interfaces use Interrupt Coalescence (IC), i.e., they generate a single interrupt for multiple packets sent or received in a short time interval. IC decreases the per-packet interrupt processing overhead. However, IC also introduces queueing delays and alters the “dispersion” (i.e., interarrival time spacing) of packet pairs or trains. In this work, we first explain how IC works in two popular Gigabit Ethernet controllers. Then, we identify the negative effects of IC on active and passive network measurements. Specifically, we show that IC can affect active bandwidth estimation techniques, causing erroneous measurements. It can also alter the packet interarrivals in passive monitors that use commodity network interfaces. Based on the “signature” that IC leaves on the dispersion and one-way delays of packet trains, we show how to detect IC and how to filter its effects from raw measurements. Finally, we show that IC can be detrimental to TCP self-clocking, causing bursty delivery of ACKs and subsequent bursty transmission of data segments. JF - Passive and Active Network Measurement EP - 256 TI - Effects of Interrupt Coalescence on Network Measurements SP - 247 KW - _2004 KW - _uanm KW - active_measurement KW - ethernet KW - experimental KW - gigabit KW - interrupt KW - measurement KW - nic KW - owd KW - passive_monitors KW - protocol KW - tcp KW - workaround AU - Prasad, Ravi AU - Jain, Manish AU - Dovrolis, Constantinos PY - 2004/// UR - http://www.springerlink.com/content/a6tyk34er2xr6bwm ER -