Assistant Professor
Interconnected Resource-aware Intelligent Systems , Eindhoven University of Technology.
My research interests are systems and networking.
Email: h dot mostafaei at tue dot nl
News
- [Mar'25] I joined the editorial board of the International Journal of Communication Systems (IJCS).
- [Feb'25] I have been promoted to Assistant Professor Level 1 (UD1). A huge and heartfelt thanks to my family, colleagues, and collaborators for their support.
- [Jan'25] "RIFO: Pushing the Efficiency of Programmable Packet Schedulers" has been accepted to the IEEE/ACM ToN.
- [Dec'24] "L3: Latency-aware Load Balancing in Multi-Cluster Service Mesh" got the Best Student Paper Award at the ACM/IFIP Middleware 2024.
- [Oct'24] "Dequeue Rate-Agnostic Switch Buffer Sharing through Packet Queueing Delay" has been accepted to the ACM CoNEXT-SW 2024.
- [Sep'24] "A Demand-aware Networked System Using Telemetry and ML with ReactNET" has been accepted to the IEEE NCA 2024.
- [Mar'24] "L3: Latency-aware Load Balancing in Multi-Cluster Service Mesh" has been accepted to the ACM/IFIP Middleware 2024.
- [Oct'23] "Per Priority Data Rate Measurement in Data Plane" has been accepted to the EuroP4 2023.
- [April'23] Delay-Resistant Geo-Distributed Analytics featured on Information Technology Daily news data analytics section.
- [Jan'23] "Real-time Link Verification in Software-Defined Networks" has been accepted to the IEEE TNSM.
- [Dec'22] Our proposal funded by the Intel ICRP’s Fast Forward Initiative (FFI ’22) for two Tofino switches.
- [Nov'22]"SDN-enabled Resource Provisioning Framework for Geo-Distributed Streaming Analytics" has been accepted to the ACM ToIT.
- [Jul'22] "Delay-Resistant Geo-Distributed Analytics" has been accepted to the IEEE TNSM.
- [Jun'22] "Network-aware Worker Placement For Wide-Area Streaming Analytics" has been accepted to the FGCS .
- [Apr'22] I joined the Eindhoven University of Technology as an Assistant Professor.
Students
| ? | Looking for one PhD candidate to work on Microservice Architecture New |
| Riz Maulana, co-supervise with N. Meratnia |
Current Research
- Programmable Networks: Programmable networks enable dynamic control over packet processing and measurement within network devices. We are working on multiple aspects of this area:
- Programmable Packet Scheduling: Programmable packet schedulers allow network operators to implement their scheduling policies on a per-packet level without changing the hardware design. Push-In-First-Out (PIFO) queues provide an abstraction for the implementation of programmable packet schedulers. However, modern programmable switches lack support for the realization of PIFO queues, and approximating PIFO behavior has the potential to overcome these limitations. We are working in several directions:
- Approximating PIFO with normalization [ToN'25]
- Network Measurement: Accurate network measurement techniques are critical for optimizing performance, detecting anomalies, and ensuring quality of service. We are exploring novel programmable measurement techniques for efficient network telemetry.
- Accurate per-priority data measurement with PrioMeter [EuroP4'23]
- Programmable Packet Scheduling: Programmable packet schedulers allow network operators to implement their scheduling policies on a per-packet level without changing the hardware design. Push-In-First-Out (PIFO) queues provide an abstraction for the implementation of programmable packet schedulers. However, modern programmable switches lack support for the realization of PIFO queues, and approximating PIFO behavior has the potential to overcome these limitations. We are working in several directions:
- Geo-Distributed Computing: Modern large-scale applications rely on geo-distributed infrastructures to achieve low-latency execution of processes while efficiently utilizing the available compute and network resources. Our research explores strategies for optimizing geo-distributed execution.
- Multi-cluster Service Mesh: Microservice architectures and service meshes have become highly popular and face increasingly stringent scalability and dependability requirements. To achieve low-latency service execution and maximize performance, service providers of large-scale distributed systems deploy microservices geographically closer to their users in multi-cluster service mesh environments. We are developing systems that can achieve the quality of service requirements of microservices.
- L3 [Middleware'24] in load balancing of microservices.
- Geo-Distributed Processing: Efficient processing of data across geographically distributed locations is critical for emerging applications. Our research focuses on network-aware resource allocation and optimization strategies.
- Multi-cluster Service Mesh: Microservice architectures and service meshes have become highly popular and face increasingly stringent scalability and dependability requirements. To achieve low-latency service execution and maximize performance, service providers of large-scale distributed systems deploy microservices geographically closer to their users in multi-cluster service mesh environments. We are developing systems that can achieve the quality of service requirements of microservices.
Teaching
| Architecture of Distributed Systems | [2022-Q1] [2023-Q1] [2024-Q1] |
| Networks, as the responsible lecturer for tutorials and essay | [2024-Q1] |
| Parallel and Distributed Computing | strating from [2025-Q4] |