Gal Mendelson
Assistant Professor
Data and Decision Sciences, Technion, Israel
galmen@technion.ac.il
Assistant Professor
Data and Decision Sciences, Technion, Israel
galmen@technion.ac.il
I am an Assistant Professor in the Faculty of Data and Decision Sciences in the Technion, Israel.
I received my PhD in Electrical Engineering from the Technion, Israel Institute of Technology on July 2020 under the supervision of Prof. Rami Atar and Prof. Isaac Keslassy. I was a Postdoc researcher in Stanford Graduate School of Business hosted by Prof. Kuang Xu.
________________________________________________________________________________________________________________
Vision: Build a Research Hub dedicated to achieving efficient resource utilization and management in energy intensive systems, with particular emphasis on data center micro-grids running AI workloads. The goal is to do more with less energy while maximizing the use of renewable resources and reducing negative impact on the environment.
________________________________________________________________________________________________________________
Research Interests: Design and management of service systems, probabilistic methods in data science and machine learning, queueing theory and stochastic processes .
Teaching Philosophy: Teaching and mentoring are among my greatest passions and one of the driving forces behind my choice of an academic career. I believe that teaching is one of the greatest impacts one can have on society and the world. I view teaching as fostering intellectual growth, practical skills acquisition, curiosity, and collaboration. I am guided by principles that include teaching through leadership, creating an inclusive environment and fostering students' engagement.
June 2025: "Load Balancing using Sparse Communication" is accepted to Operations Research!
This work with Kuang Xu (Stanford GSB) shows that it is possible to obtain good load balancing performance while reducing 90% of the communication between servers and the load balancer.
July 2024: Recipient of Technion's Excellence in Teaching award (Discrete Mathematics, ~250 students)
It has been a privilege to teach this fundamental undergraduate course. I had the opportunity to facilitate a meaningful learning process of fundamental mathematical thinking and tools (e.g., logic, set theory and "what is a proof?"). This has been extra challenging (for me and them) due to the horrendous events at October 7th and the ongoing war. I am grateful to the students for taking this difficult, but rewarding journey with me.
June 2023: Honored to have served in the judging committee of the Informs Applied Probability Best Student Paper award!
January 2023: I joined the Faculty of Data and Decision Sciences in the Technion, Israel!
June 2022: Participated in the full time Stanford Graduate School of Business Ignite entrepreneurship program!
June 2022: Honored to have served in the judging committee of the Informs Applied Probability Best Student Paper award!
Load Balancing Using Sparse Communication (arXiv), Accepted to Operations Research, 2025
Gal Mendelson, Xu Kuang
Detecting Service Slowdown Using Observational Data (arXiv), preprint, 2024 (under revision)
Gal Mendelson, Xu Kuang
The Hybrid Hospital: Balancing On-Site and Remote Hospitalization (arXiv) , 2024 (under revision)
Noa Zychlinski, Gal Mendelson and Andrew Daw
Worst Case Performance in Piece-wise Stationary Multi-Armed-Bandits, in progress, 2024
Gal Mendelson, Eyal Tadmor (Undergraduate student)
Trap Sampling: a New Method of Sampling Data Streams, in progress, 2024
Gal Mendelson, Ilai Avni (M.Sc. student)
Approximating Random Rotations using Hadamard transforms, in progress, 2024
Gal Mendelson, Tomer Zilca (M.Sc. student)
EDEN: Communication-Efficient Federated Learning via Robust Distributed Mean Estimation (arXiv), Accepted to ICML, 2022
Shay Vargaftik, Ran Ben Basat, Amit Portnoy, Gal Mendelson, Yaniv Ben-Itzhak, Michael Mitzenmacher
Load Balancing with JET: Just Enough Tracking for Connection Consistency, Accepted to CoNEXT, 2021
Gal Mendelson, Shay Vargaftik, Katherine Barabash, Dean Lorentz, Isaac Keslassy and Ariel Orda
DRIVE: One-bit Distributed Mean Estimation, Accepted to NeurIPS, 2021
Shay Vargaftik, Ran Ben Basat, Amit Portnoy, Gal Mendelson, Yaniv Ben-Itzhak, Michael Mitzenmacher
On the Persistent-Idle Load Distribution Policy Under Batch Arrivals and Random Service Capacity (link), Preprint
Rami Atar, Isaac Keslassy, Gal Mendelson, Ariel Orda and Shay Vargaftik
A Lower Bound on the Stability Region of Redundancy-d with FIFO Service Discipline (link), Operations Research Letters, 2021
Gal Mendelson
AnchorHash: A Scalable Consistent Hash (link), IEEE Transactions on Networking, 2020
Gal Mendelson, Shay Vargaftik, Katherine Barabash, Dean Lorentz, Isaac Keslassy and Ariel Orda
Persistent-Idle Load-Distribution, Stochastic Systems, 2020
Rami Atar, Isaac Keslassy, Gal Mendelson, Ariel Orda and Shay Vargaftik
Sub-Diffusive Load-Balancing in Time-Varying Queueing Systems (link), Operations Research, 2019
Rami Atar, Isaac Keslassy and Gal Mendelson
Replicate to the Shortest Queue (link), Queueing Systems, 2019
Rami Atar, Isaac Keslassy and Gal Mendelson
drmt: Disaggregated Programable Switching (link), SIGCOMM, 2017
Sharad Chole, Andy Fingerhut, Sha Ma, Anirudh Sivaraman, Shay Vargaftik, Alon Berger, Gal Mendelson, Mohammad Alizadeh, Shang-Tse Chuang, Isaac Keslassy, Ariel Orda and Tom Edsall
On the Non-Markovian Multiclass Queue with Risk Sensitive Cost (link), Queueing Systems, 2016
Rami Atar, Gal Mendelson
2024: Exceptional Teaching Award (Discrete Mathematics)
2020: Fulbright Postdoctoral Scholarship
2020: Honorable mention: The George B. Dantzig Dissertation award, INFORMS
2019: Applied Probability Society best student paper award for "Sub-Diffusive Load-Balancing in Time-Varying Queueing Systems", INFORMS
2019: The Joseph Perl prize for excellent research in communication systems
2017-2020: Hasso Plattner Institute Ph.D. Scholarship
2008-9, 2015, 2020: Excellent Teaching Award
Committees
Judging committee of the Informs Applied Probability Best Student Paper Award (2022,2023)
Review (quantity indicated in (x))
Mathematics of Operations Research (4), Operations Research (3), Management Science (1), Queueing Systems (2), European Journal of Operations Research (1), ACM Transactions on Internet of Things (1), Israeli Science Foundation (ISF) Grant (1)
Conferences
Session chair: Informs Annual Meeting (4), Applied Probability Society (APS) conference (1).
Current:
Spring 2024-2025: Service Engineering, Data and Decision Sciences, Technion
Spring 2024-2025: Discrete Mathematics, Data and Decision Sciences, Technion
Recent:
Winter 2023-2024, Winter 2024-2025: Discrete Mathematics, Data and Decision Sciences, Technion
Spring 2022-2023, Spring 2023-2024: Service Engineering, Data and Decision Sciences, Technion
Past:
Graduate level courses:
2017, 2018: Stochastic processes and applications, Industrial Engineering and Management, Technion
2016, 2017: Foundations of stochastic processes, Electrical Engineering, Technion
Undergraduate level courses:
2016-2019: Random signals, Electrical Engineering, Technion
2016-2019: Introductory project in EE, Electrical Engineering, Technion
2008-9, 2015: Probability, Algebra 1, Calculus 1 and 2, Ordinary diff. eq., Partial diff. eq., Mathematics, Technion
Current:
Dr. Ankita Sen, Postdoc, Online Decision Making in Inventory Problems
Ilai Avni, M.Sc, Trap Sampling: a New Method of Sampling Data Streams
Tomer Zilca, M.Sc, Approximating Random Rotations using Hadamard transforms
Eyal Tadmor, Undergraduate, Worst Case Performance in Piece-wise Stationary Multi-Armed-Bandits
Past students:
Yossi Bachrach, Undergraduate
"Connecting PetConnect": A Digital Transformation of PetConnect's Workflow.
PetConnect is a non-profit dedicated to rescuing and rehoming pets. The students developed a new web system platform which centralizes data, streamlines management, provides analytics, matches adopters with dogs, and generates personalized shopping lists, aiming to significantly boost efficiency and save time.
Nour Eldne Shannan, Or Haziza, Yuval Dror, Katia Shammas, Noa Rahamimov
"OPT Tov": Increasing throughput in a frozen product packaging line
The students designed solutions for improving the packaging process in a factory that produces packaged frozen products, based on understanding the factory's needs and goals, collecting data and analyzing the effect of their proposed solutions.
Florian Tordjman, David Poignon Cahen, Naomie Melloul, Nathaniel Adda
2020: Nvidia (Mellanox), Israel, researcher. Scheduling in optical circuit switching networks and routing in expander based topologies
2018: IBM research, Israel, internship. Using alarm data to predict network failures
2017: IBM research, Israel, internship. Hash based data center load balancing