Estimate CPU Utilization for Data Processing with DPDK

dc.contributor.authorRydén, Anton
dc.contributor.authorNäslund, Erik
dc.contributor.departmentChalmers tekniska högskola / Institutionen för data och informationstekniksv
dc.contributor.examinerSvensson, Lars
dc.contributor.supervisorStrikos, Panagiotis
dc.date.accessioned2022-06-23T09:04:43Z
dc.date.available2022-06-23T09:04:43Z
dc.date.issued2022sv
dc.date.submitted2020
dc.description.abstractIn a world where networking and network systems are becoming more and more important, it is crucial that these systems are properly developed and monitored. The purpose of the project and aim is to create an algorithm that estimates the CPU workload during packet processing that is linear with respect to network traffic using the Data Plane Development kit (DPDK). With such a tool, it becomes possible to predict when a system might start dropping packets. An important part of the project is the design flow, which mainly consists of a single machine with NUMA design along with tools to gather and visualize the results. A challenge of this project is a lack of prior attempts, making it difficult to find a reference point. While a linear solution was found, it is not without its flaws. However, with further testing, a practical implementation can be achieved. One conclusion to the project is that it can serve as a baseline for future attempts and research.sv
dc.identifier.coursecodeLMTX38sv
dc.identifier.urihttps://hdl.handle.net/20.500.12380/304882
dc.language.isoengsv
dc.setspec.uppsokTechnology
dc.subjectDPDKsv
dc.subjectdata processingsv
dc.subjectload estimationsv
dc.subjectTRexsv
dc.subjectpacket dropssv
dc.titleEstimate CPU Utilization for Data Processing with DPDKsv
dc.type.degreeExamensarbete på kandidatnivåsv
dc.type.uppsokM2

Ladda ner

Original bundle

Visar 1 - 1 av 1
Hämtar...
Bild (thumbnail)
Namn:
22-17 Ryden Näslund.pdf
Storlek:
2 MB
Format:
Adobe Portable Document Format
Beskrivning:

License bundle

Visar 1 - 1 av 1
Hämtar...
Bild (thumbnail)
Namn:
license.txt
Storlek:
1.51 KB
Format:
Item-specific license agreed upon to submission
Beskrivning: