netmap: A Novel Framework for Fast Packet I/O

By Luigi Rizzo. USENIX Annual Technical Conference, 2012

943 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/6491e05237887e4d6c4f7df1a8aecad445bca34f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

MoonGen: A Scriptable High-Speed Packet Generator (2014, background): “Network stacks of operating systems come with a high overhead [17].”
Design Guidelines for High Performance RDMA Systems (2016, background): “Low-level factors in network I/O Although there is a large body of work that measures the throughput and CPU utilization of network communication [18, 16, 26, 13, 20], there is less existing literature on understanding the lowlevel behavior of network cards.”
Shenango: Achieving High CPU Efficiency for Latency-sensitive Datacenter Workloads (2019, background): “Kernel-bypass networking: Many systems bypass the kernel to achieve low-latency networking by using RDMA, SR-IOV, or libraries such as DPDK [2] or netmap [66].”
Maglev: A Fast and Reliable Software Network Load Balancer (2016, background): “It has sustained the rapid global growth of Google services, and it also provides network load balancing for Google Cloud Platform.”
Packet-Level Telemetry in Large Datacenter Networks (2015, methodology): “Given 100 Tbps of DCN trafﬁc, 4 Advanced packet capturing libraries like DPDK [1] NetMap [32] and WireCap [34] can support more than 20Mpps using 8 cores. the mirrored trafﬁc requires 380 Gbps network bandwidth and 300M pps CPU processing power.”
The QUIC Transport Protocol: Design and Internet-Scale Deployment (2017, background): “Specifically, general kernel bypass [57] seems like a promising match for a user-space transport.”
The eXpress data path: fast programmable packet processing in the operating system kernel (2018, methodology): “This operating mode is quite similar to the approach used by frameworks such as Netmap [46] and PF_RING [11], which offer high packet processing performance by lowering the overhead of transporting packet data from the network device to a userspace application, without bypassing the kernel completel”
Planck (2014, methodology): “Each monitoring port is connected to a collector running on a separate server that uses netmap [30] to process traffic sent by a monitoring port at line-rate.”
SketchVisor: Robust Network Measurement for Software Packet Processing (2017, background): “Extensive work has been undertaken to improve software-based packet forwarding performance [16, 45].”
Congestion Control for Large-Scale RDMA Deployments (2015, methodology): “TCP’s latency can potentially also be reduced using technologies such as Netmap [28], Intel Data Direct I/O [45] and mTCP [23].”
The Design and Implementation of Open vSwitch (2015, unknown): “To this end, there is an ongoing effort to add both DPDK [11] and netmap [30] support to Open vSwitch.”
A comprehensive survey of Network Function Virtualization (2018, methodology): “[47] L. Rizzo, Netmap: a novel framework for fast packet i/o, in: 2012 USENIX An- nual Technical Conference, Boston, MA, 2012, doi: 10.1145/774763.774771 .”