Datacenter computing collaboration
As part of a larger research collaboration on datacenter computing between the SWYSTEMS group at USI and SAP's Innovation Office for Global Cloud Services we are seeking qualified applicants for several open positions both at the PhD student and postdoctoral researcher levels. This broad academic/industrial research collaboration yields exciting opportunities for cutting edge systems research with the ability of practical validation, and is also backed by the Swiss National Science Foundation and other funding organs.
USI: The Università della Svizzera italiana (USI) is a rising star (as shown, e.g., by its continuous high ranking among young universities
), notably in the field of computer science
, and hosts one of the three largest departments in the field in Switzerland with over 50 professors and lecturers. USI is located in one of the most beautiful locations in the country, and is well-connected through its own domestic airport, and accessible from Milano airport by bus.
SWYSTEMS: The Software Systems (SWYSTEMS) group led by Prof. Patrick Eugster
is part of the newly founded Computer Systems Institute
at USI. The group looks back at 20 years of international experience in cutting edge research in collaboration with corporate partners. The group's philosophy is to focus on highly relevant hard problems and choose (top-tier) publication venues accordingly, rather than vice-versa.
is Europe's largest software house and largest cloud service provider. The set-up of the Innovation Office of Global Cloud Services (GCS) is highly interdisciplinary to understand the current state of the art, challenges, and trade-offs in datacenter designs, and to find opportunities to improve datacenters by tailoring designs to the demands of SAP's cloud offerings.
Official start is ASAP, but strong candidates who wish to start later this year are encouraged to apply now as well. Positions are fully funded with highly competitive salaries, starting at ~80K CHF and ~47K CHF yearly gross for postdoctoral and PhD levels respectively.
Research areas and topics
The project involves a number of topics in the context of datacenter computing, including, but not limited to [with some recent sample publications, see bottom for details]:
- Networks, e.g., monitoring and management, congestion control , traffic engineering , packet scheduling [3,4], service function chains , in-network computing 
- Distributed systems, e.g., resource management , asynchrony/synchrony trade-offs , fault-tolerant agreement and coordination [7,8,9], distributed data processing and storage [10,11],
- Security, e.g., network security [8,9], secure communication , privacy-preserving data processing 
- Programming languages, e.g., domain-specific languages for network programming , elastic application programming [14,15], verification of distributed systems [16,17,18]
Application procedure and requirements
The ideal candidate has strong background in one or several of the areas mentioned above depending on the topic(s) of interest.
For the postdoctoral level, background must be demonstrated by a PhD thesis completed (or close to completion) in a relevant area, and publications in respective top tier venues. Applicants at the PhD level must hold a prior degree in computer science/engineering.
In addition all applicants must be proficient in spoken and written scientific English, have excellent communication skills, be creative, and possess leadership qualities.
Candidates are requested to submit their applications via email with subject header "DC computing position application" to [email protected]
indicating their field(s) of interest.
Applications must include a 1-page research summary, CV, and names of 2 or 3 references for PhD and postdoc levels respectively. Screening will start immediately and continue until current positions are filled, with additional positions becoming available throughout the year.
Prior group members have moved on to further exciting opportunities
, several notably at SAP.
] P. Taheri, D. Menikkumbura, E. Vanini, S. Fahmy, P. Eugster, and T. Edsall. RoCC: Robust Congestion Control
. ACM CoNEXT'20
] P. Jahnke, E. Stapf, J. Mieseler, G. Neumann, and P. Eugster. Towards Fine Grained Network Flow Prediction
. CoRR abs/1808.06453, 2018
] P. Eugster, A. Kesselman, K. Kogan, S. Nikolenko, and A. Sirotkin. Admission Control in Shared Memory Switches
. Journal of Scheduling 21(5), 2018
] K. Kogan, D. Menikkumbura, G. Petri, Y. Noh, S. Nikolenko, A. Sirotkin, and P. Eugster. Towards Software-Defined Buffer Management
. IEEE/ACM Transactions on Networking 28(5), 2020
] M. Blöcher, R. Khalili, L. Wang, and P. Eugster. Letting off STEAM: Distributed Runtime Traffic Scheduling for Service Function Chaining
. IEEE INFOCOM'20
] M. Blöcher, L. Wang, P. Eugster, and M. Schmidt. Switches for HIRE: Resource Scheduling for Datacenter In-Network Computing
. ACM ASPLOS'21
] P. Jahnke, P.-L. Roman, P. Chuprikov, and P. Eugster. Live in the Express Lane
. USENIX ATC'21
] J. Lembke, S. Ravi, P.-L. Roman, and P. Eugster. Consistent and Secure Network Updates made Practical
. ACM/IFIP Middleware'20
] J. Lembke, S. Ravi, P. Eugster, and S. Schmid. RoSCo: Robust Updates for Software-defined Networks
. IEEE Journal on Selected Areas in Communication 38(7), 2020
] M. Blöcher, E. Coppa, P. Kleber, P. Eugster, W. Culhane, and M. Saeida Ardekani. ROME: All Overlays Lead to Aggregation, but Some Are Faster than Others
. ACM Transactions on Computer Systems, 2022 (to appear)
] J. Lembke, P.-L. Roman, and P. Eugster. DEFUSE: An Interface for Fast and Correct User-Space File-System Access
. ACM Transactions on Storage, 2022 (to appear)
] P. Eugster, G. A. Marson, and B. Poettering. A Cryptographic Look at Multi-party Channels
. IEEE CSF'18
] S. Savvides, D. Khandelwal, and P. Eugster. Efficient Confidentiality-Preserving Data Analytics over Symmetrically Encrypted Datasets
. VLDB Endowment 13(8), 2020
] B. Sang, P.-L. Roman, P. Eugster, H. Lu, S. Ravi, and G. Petri. PLASMA: Programmable Elasticity for Stateful Cloud Computing Applications
. ACM EuroSys'20
] B. Sang, P. Eugster, G. Petri, S. Ravi, and P.-L. Roman. Scalable and Serializable Networked Multi-Actor Programming
. ACM OOPSLA'20
] M. Viering, T.-C. Chen, P. Eugster, R. Hu, and L. Ziarek. A Typing Discipline for Statically Verified Crash Failure Handling in Distributed Systems
] M. Viering, R. Hu, P. Eugster, and L. Ziarek. A Multiparty Session Typing Discipline for Fault-tolerant Event-driven Distributed Programming
. ACM OOPSLA'21
] M. Marescotti, R. Otoni, L. Alt, P. Eugster, A. Hyvärinen, and N. Sharygina. Accurate Smart Contract Verification Through Direct Modelling