Obtained while pursuing a Ph.D. degree as a post-bachelor
Thesis: "Effective Cache Bank Placement for GPUs"
Advisor: Prof. Hamid
Sarbazi-Azad
Affiliated with the National Organization for Development of Exceptional Talents (NODET).
Best Paper Award in the Real-Time and Embedded Technology and Applications Symposium (RTAS) 2020
Ranked 285th (top 0.1%) in the National University Entrance Examination, Among more than 380,000 participants, Iran, 2012
Member of National Organization for Development of Exceptional Talents (NODET), 2005 - 2012
Selected to study in Shahid Dastgheib High school, Through an exam with less than 1% acceptance rate. Sept. 2008
Semifinalist in 27th, 28th, 29th Iranian National Olympiad in Mathematics 2009, 2010, 2011
Semifinalist in 19th, 20th Iranian National Olympiad in Informatics (INOI) 2009, 2010
{Shahin Roozkhosh, Bassel El Mabsout}, Siddharth Mysore, Kate Saenko, Renato Mancuso. SwaNNFlight and Anchored Learning for On-the-Fly Sim-to-Real Adaptation. Under Review.
Weifan Chen, Ivan Izhbirdeev, Denis Hoornaert, Shahin Roozkhosh, Sanskriti Sharma, Patrick Carpanedo and Renato Mancuso. Timely Progress Integrity: Low-overhead Online Assessment of Timely Progress as a Commodity. Under Review.
Renato Mancuso, Shahin Roozkhosh, Denis Hoornaert, Ju Hyoung Mun, Tarikul Islam Papon, Manos Athanassoulis. RSoftware-Shaped Platforms. Real-time And intelliGent Edge computing workshop (RAGE), RAGE, 2023.
Shahin Roozkhosh, Denis Hoornaert, Ju Hyoung Mun, Tarikul Islam Papon, Ahmed Sanaullah, Ulrich Drepper, Renato Mancuso, Manos Athanassoulis. Relational Memory: Native In-Memory Accesses on Rows and Columns. In 26th International Conference on Extending Database Technology (EDBT), IEEE, 2023.
Tarikul Islam Papon, Ju Hyoung Mun, Shahin Roozkhosh, Denis Hoornaert, Ahmed Sanaullah, Ulrich Drepper, Renato Mancuso, Manos Athanassoulis. Relational Fabric: Transparent Data Transformation. In 39th IEEE International Conference on Data Engineering (ICDE), Under Review. ICDE, 2023.
Shahin Roozkhosh, Denis Hoornaert, Renato Mancuso. CAESAR: Coherence-Aided Elective and Seamless Alternative Routing via on-chip FPGA. In 43rd IEEE Real-Time Systems Symposium. (RTSS), IEEE, 2022.
Mattia Nicolella, Denis Hoornaert, Shahin Roozkhosh, Andrea Bastoni, Renato Mancuso. Know your Enemy: Benchmarking and Experimenting with Insight as a Goal. In 43rd IEEE Real-Time Systems Symposium (RTSS@work), IEEE, 2022.
Shahin Roozkhosh, Denis Hoornaert, and Renato Mancuso. Hardware Data Re-organization Engine for Real-Time Systems. In 43rd IEEE Real-Time Systems Symposium (RTSS@work), IEEE, 2022.
Mattia Nicolella, Shahin Roozkhosh, Denis Hoornaert, Andrea Bastoni, Renato Mancuso. RT-Bench: an Extensible Benchmark Framework for the Analysis and Management of Real-Time Applications. The 30th International Conference on Real-Time Networks and Systems (RTNS), 2022.
{Shahin Roozkhosh, Denis Hoornaert}, Renato Mancuso. A Memory Scheduling Infrastructure for Multi-core Systems with Re-programmable Logic. In 33rd Euromicro Conference on Real-Time Systems (ECRTS), IEEE, 2021
Denis Hoornaert, Shahin Roozkhosh, Renato Mancuso and Marco Caccamo. Identifying Unexpected Inter-coreInterference Induced by Shared Cache. In 27th IEEE Real-Time and Embedded Technology and Applications Symposium (WiP session) (RTAS), IEEE, 2021.
Dharmesh Tarapore, Shahin Roozkhosh, Steven Brzozowski, Renato Mancuso. Observing the Invisible: Live Cache Inspection for High-Performance Embedded Systems. In IEEE Transactions on Computers (IEEE TC), IEEE, 2021.
(Best Paper Award) Shahin Roozkhosh, Renato Mancuso. The potential of programmable logic in the middle: cache bleaching. In 26th Real-Time and Embedded Technology and Applications Symposium (RTASS), , pp. 296-309. IEEE, 2020.
Sadrosadati, Mohammad, Ramin Bashizade, Shahin Roozkhosh, Ali Shafiee, and Hamid Sarbazi-Azad. "A Method to Improve Adaptivity of Odd-Even Routing Algorithm in Mesh NoCs." In Parallel, Distributed, and Network-Based Processing (PDP), 2016 24th Euromicro International Conference on, pp. 755-758. IEEE, 2016.
Sadrosadati, Mohammad, Amirhossein Mirhosseini, Shahin Roozkhosh, Hazhir Bakhishi, and Hamid Sarbazi-Azad. "Effective Cache Bank Placement for GPUs" In 2017 Design, Automation Test in Europe Conference Exhibition (DATE), pp. 31-36. IEEE, 2017.
Affiliated Researcher, Supervisor: Ulrich Drepper, Red Hat Research
Advisors: Prof. Manos Athanassoulis and Prof. Renato Mancuso
I designed and implemented a Re-configurable Hardware Accelerator. A novel memory abstraction that enables the definition of workload-specific memory access paradigms through an on-chip FPGA.
The proposed accelerator now available open-source (here)
Research Assistant, Supervisor: Prof.
Hamid Sarbazi-Azad, School of Computer
Science
Advisor: Dr. Arash Tavakkol and Mohammad
Sadrosadati
We proposed a Performance Evaluation of Dynamic Page Allocation Strategies in
SSDs.
Corresponding Paper Published in ACM Transactions on Modeling and Performance Evaluation
of Computing Systems (TOMPECS)
Xmulator is an object-oriented event-based simulator software for interconnection networks and wireless networks. I contributed to the packages required for Network-on-Chip (NoCS) simulation. Xmulator uses Orion power library for power and energy estimation.
DiskSim is an effcient, accurate, highly-con gurable disk system simulator which includes modules for most secondary storage components of interest, including device drivers, buses, controllers, adapters, and disk drives.
Research Assistant, Supervisor: Prof. Hamid Sarbazi-Azad, Department of Computer Engineering, Sharif University of Technology.
The placement of the Last Level Cache (LLC) banks in the GPU on-chip network can significantly affect the performance of memory-intensive workloads. We attempt to offer a placement methodology for the LLC banks to maximize the performance of the on-chip network connect- ing the LLC banks to the streaming multiprocessors in GPUs.
GPGPU-Sim provides a detailed simulation model of a contemporary GPU (such as NVIDIA's Fermi and GT200 architectures) running CUDA and/or OpenCL workloads and now includes an integrated (and validated) energy model, GPUWattch.
The software was linked to GPGPU-Sim to process all data collected from previous
simulations
automatically and lead us to nd a new throughput aware metric in The placement of the
Last
Level Cache (LLC) banks in the GPU on-chip network.
Genetic Algorithm (GA) is
a
metaheuristic inspired by the process of natural selection that
belongs to the larger class of evolutionary algorithms (EA).
We figured a novel approach, called Preemptive Waiting, which applied to Odd-Even routing
algorithm (PWOE). PWOE postpones the saturation traffic rate of NoC compared to OE,
under
synthetic traffic loads.
BookSim
which is a cycle-accurate simulator developed in C++ was
as our Simulation Environment.
Full-time engineer to Shell International Exploration and Production Company (SIEP), of which Shell TechWorks is a part. Responsibilities include but are not limited to developing or directing embedded system testing or validation procedures and embedded programming.
Appetizer is an Integrated Management System for Food Services and Clients which inform them about features like Check-in, Reserve, Takeaway and more Services
Peeyade is a media application that gives users location-based information about Tehran
Job Description: Developing Android mobile applications and Implementing back-end services using PHP.
Digital System Design - Fall 2015
Automata and Compiler - Spring 2015
Computer Structure and Language - Fall 2015
Discrete Structures - Fall 2015
Logic Design - Fall 2014
Computer Architecture - Fall 2014
Advanced Programming - Fall 2013
Fundamentals of Programming - Fall 2013, Spring 2013
Private C and C++ Programming Tutor 2013 - Now
Private English Tutor 2016 - Now
With a concentration in Portrait and Documentary Photography.
Photography is also my
avocation.
I love the feeling of sharing my experiences with others through my blog.