At Georgia Tech, my research interests evolve around the area of Operating Systems, focusing on the memory management layer. More specifically, I work on designing and building methods for optimizing application performance in systems with heterogeneous memory components.
Prior to Georgia Tech, I received an undegraduate Diploma on ECE, with major in Computer Science, at the School of Electrical and Computer Engineering at National Technical University of Athens in Greece.
This student profile featured in GT SCS news nicely summarizes my academic journey.
Unexpected Performance of Intel Optane DC Persistent Memory. Tony Mason, Thaleia Dimitra Doudali, Margo Seltzer, Ada Gavrilovska. In IEEE Computer Architecture Letters, vol.19, no.1, pp.55-58, 1 Jan.-June 2020. [IEEE link]
Kleio: a Hybrid Memory Page Scheduler with Machine Intelligence. Thaleia Dimitra Doudali, Sergey Blagodurov, Abhinav Vishnu, Sudhanva Gurumurthi and Ada Gavrilovska. In Proceedings of the 28th International Symposium on High-Performance Parallel and Distributed Computing (HPDC '19), Phoenix, AZ, USA, June 22 - 29 2019.
Best Paper Award Finalist. Best PhD Forum Poster with Research Paper Award.
[ACM link] [slides] [poster] [award] [GT SCS News]
Mnemo: Boosting Memory Cost Efficiency in Hybrid Memory Systems. Thaleia Dimitra Doudali and Ada Gavrilovska. In proceedings of the 5th IEEE International Workshop on High-Performance Big Data, Deep Learning, and Cloud Computing (HPBDC 2019). In conjunction with The 33rd IEEE International Parallel and Distributed Processing Symposium (IPDPS 2019). Rio de Janeiro, Brazil, May 2019. [paper] [slides] [poster] [GitHub]
CoMerge: Toward Efficient Data Placement in Shared Heterogeneous Memory Systems. Thaleia Dimitra Doudali and Ada Gavrilovska. In Proceedings of MEMSYS 2017, Alexandria, VA, USA, October 2–5, 2017, 11 pages. [ACM link] [slides] [poster] [GitHub]
Spaten: a Spatio-temporal and Textual Big Data Generator. Thaleia Dimitra Doudali, Ioannis Konstantinou, Nectarios Koziris. In proceedings of the 2017 Big Spatial Data Workshop (BSD 2017 in conjuction with IEEE BigData 2017), Boston, MA, USA December 11-14 2017, 6 pages. [paper] [slides] [GitHub]
Design and implementation of Spaten, a configurable generator of spatio-temporal and textual social media data. Scalability testing of an HBase cluster that stores a large dataset produced by the generator. [pdf] [slides]
- [February 2020] Kleio: a Hybrid Memory Page Scheduler with Machine Intelligence. At the Grascomp/EuroSys Shadow PC joint workshop at UC Louvain, Belgium.
- [January 2020] Kleio: a Hybrid Memory Page Scheduler with Machine Intelligence. At the 5th Computing Systems Research Day at NTUA, Greece.
Program Committee Service: EuroSys '20 Shadow PC.
Student Travel Grant Awards: Usenix ATC '20, TCPP award for IPDPS '19, SoCC '18, Usenix ATC '17, OSDI '16.
Coneferences Attended: Usenix ATC '20, HPDC '20, EuroSys '20, HPDC '19, IPDPS '19, Eurosys '19, SoCC '18, Salishan '18, IEEE Big Data '17, MEMSYS '17, Usenix ATC '17, VMWare Radio '17, CRA-W Grad Cohort '17, OSDI '16, ASPLOS '15.
[EuroDW '20 ] Intelligent and Cost-Efficient Data Management for Hybrid Memory Systems.
[IPDPS PhD Forum ‘19] Network-aware Data Management for Disaggregated Memory Systems. [poster]
[Eurosys ‘19] Page Scheduling on Hybrid Memory Systems with Machine Intelligence. [poster]
[SoCC ‘18] Mnemo: Boosting Memory Cost Efficiency in Hybrid Memory Systems. [poster]
[Salishan ‘18] Data Management in Heterogeneous Memory Systems. [poster]
[CRA-W Grad Cohort ‘17] MemSpan: Toward Efficient Data Placement in Heterogeneous Memory Systems. [poster]
PhD in Computer Science.
Research area: Operating Systems.
Advisor: Ada Gavrilovska
Teaching Assistanships: Spring 2017 and 2019 for Advanced Operating Systems (CS 6210).
Intern at AMD Research - Bellevue WA.
Worked on the design of data scheduling policies with machine intelligence on hybrid memory systems.
Patents submitted! USPTO Application No. 16/220,462 and 16/425,878
Mentor: Sergey Blagodurov
Manager: Andrew Kegel
Intern at the CPBU Palo Alto.
Distributed Resource Scheduler Team.
Worked on the design of techniques for container scheduling in a virtualized environment.
Patent submitted! USPTO Application No. 16/034,126
Mentor: Zhelong Pan
Manager: Lan Gao
Intern at the Data Domain File System - Garbage Collection Team.
Worked on developing techniques that can group similar files, implementing clustering algorithms and similarity estimations on a deduplication dataset.
Mentor: Abhinav Duggal
Manager: Aseem Vaid
Diploma in ECE. Major in Computer Science.
Thesis Advisor: Nectarios Koziris
Thesis Mentor: Ioannis Konstantinou
[USPTO Application No. 16/425,878] Computer Resource Scheduling Using Generative Adversarial Networks.
Sergey Blagodurov, Abhinav Vishnu, Thaleia Dimitra Doudali, Jagadish B. Kotra
[USPTO Application No. 16/220,462] Multi-instance Recurrent Neural Network Prediction.
Sergey Blagodurov, Thaleia Dimitra Doudali, Amin Farmahini Farahani
[USPTO Application No. 16/034,126] Techniques for Container Scheduling in a Virtual Environment.
Thaleia Dimitra Doudali, Zhelong Pan, Pranshu Jain