• No results found

Master Thesis – High performance scheduling for packet processing in 5G

N/A
N/A
Protected

Academic year: 2021

Share "Master Thesis – High performance scheduling for packet processing in 5G"

Copied!
1
0
0

Loading.... (view fulltext now)

Full text

(1)

Contact Person

Johan Wibeck +46 10 711 40 06

johan.wibeck@ericsson.com

Master Thesis –

High performance scheduling for packet processing in 5G

Background

Mobile networks are used all over the world and are the corner stone for the networked society, where everything shall be connected. To support the vast amount and diversity of data expected in future networks, Ericsson are developing products to drive and support the networked society. The subjects for Master Thesis are defined to investigate and develop algorithms, architecture, tools etc. to support huge increase of speech, data and massive IoT for Radio Access Networks.

Thesis Description

Packet processing on general purpose CPUs is a strong industry trend. In order to make it possible to create flexible implementations it would be beneficial to view the problem as a set of scheduled stages in a packet pipe line. This puts very high requirements on a scheduler to move packets between stages. DPDK is a set of open source libraries and drivers for fast packet processing on generic hardware.Implement and test a DPDK EventDev scheduler using statistical heuristics for high performance scheduling.

Task:

• Study and evaluate possible packet schedulers for a multi stage packet processing pipeline written on top of DPDK.

The thesis will be concluded with a result presentation for the Ericsson team.

Qualifications

This project aims at students in electrical engineering, computer science, computer engineering or similar.

Background in wireless communication is preferred.

Extent

1-2 students, 30hp each

Location

Ericsson AB Mjärdevi, Linköping

Preferred Starting Date

Spring 2018

Keywords

Mobile Telecommunication, Optimization, Packet Processing, DPDK, 5G

References

Related documents

This thesis work will investigate machine learning methods for improving 5G network performance by automatically tuning key network parameters to their optimal values..

This thesis will investigate the handover procedure and related handover parameter optimization for different types of performance enhancements in a 5G network.. For example,

Placing multiple antennas at the transmitter and using a single receive antenna creates a multiple-input-single-output (MISO) channel. Multiple signals are now transmitted, instead

The result for the vEPG deployment we discussed in section 3.4.3 with both the user plane VMs on the first NUMA node and the control plane VMs on the second NUMA node is shown in

Nu utvidgar vi ARIMA-modellen med att också lägga in en förklarande variabel. Modellen tillåts dessutom att innehålla flera laggar av den förklarande variabeln. Först plottar

Den rekommenderade metoden Simpsons biplan jämförs med Philips automatiska funktion a2Dq, som utgår ifrån speckle tracking principen, för bedömning av ejektionsfraktionen..

For example, epitaxial layers grown on 6H- SiC at the same temperature and coil position, but different insulation setups exhibit significantly larger surface coverage with

These are an empirical study of crisis volunteerism (CV) during and after the Swedish forest fires crisis in 2018, and a literature study of previous research on information