-
Notifications
You must be signed in to change notification settings - Fork 9
/
CITATION.cff
52 lines (52 loc) · 1.79 KB
/
CITATION.cff
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
cff-version: 1.2.0
title: 'PacketMill: Toward per-core 100-Gbps Networking'
message: >-
If you use PacketMill or X-Change in any context,
please cite our repository.
type: software
authors:
- given-names: Alireza
email: [email protected]
affiliation: KTH Royal Institute of Technology
orcid: 'https://orcid.org/0000-0001-5083-4052'
family-names: Farshin
- given-names: Tom
family-names: Barbette
email: [email protected]
affiliation: KTH Royal Institute of Technology
orcid: 'https://orcid.org/0000-0003-1269-2190'
identifiers:
- type: doi
value: 10.1145/3445814.3446724
description: PacketMill paper
- type: doi
value: 10.5281/zenodo.4435970
description: PacketMill Artifact (ASPLOS'22)
repository-code: 'https://github.com/aliireza/packetmill'
url: 'https://packetmill.io/'
repository: 'https://zenodo.org/record/4435970#.YfkrFVvMKV5'
abstract: >-
We present PacketMill, a system for optimizing
software packet processing, which (i) introduces a
new model to efficiently manage packet metadata and
(ii) employs code-optimization techniques to better
utilize commodity hardware. PacketMill grinds the
whole packet processing stack, from the high-level
network function configuration file to the
low-level userspace network (specifically DPDK)
drivers, to mitigate inefficiencies and produce a
customized binary for a given network function. Our
evaluation results show that PacketMill increases
throughput (up to 36.4 Gbps -- 70%) & reduces
latency (up to 101 us -- 28%) and enables
nontrivial packet processing (e.g., router) at ~100
Gbps, when new packets arrive >10× faster than main
memory access times, while using only one
processing core.
keywords:
- PacketMill
- DPDK
- FastClick
- X-Change
- Metadata Management
license: MIT