HAR.S.H. (Hardware-Aware extReme-scale Similarity search)

We are delighted to announce the first confirmed speakers for SPTDC 2026, along with our sponsors: - Giuliano Losa - Stellar Development Foundation, USA - Alberto Sonnino - Mysten Labs and University College London, UK - Nirupam Gupta - University of Copenhagen, Denmark - Naama Ben-David - Technion – Israel Institute of Technology, Israel - Sara Tucci Piergiovanni - University of Paris-Saclay, France - Sonia Ben Mokhtar - INSA Lyon, France - Siddhartha Jayanti - Dartmouth College, USA - Thomas Nowak - ENS Paris-Saclay, France - Gadi Taubenfeld - Reichman University, Israel - Fabien Dufoulon - Lancaster University, UK When : April 27 - 30, 2026  Where : Heraklion, Crete, Greece. Website  : https://lnkd.in/dE-uC4BE SPTDC 2026 will take place in the historic city of Heraklion on the island of Crete and will bring together leading researchers and practitioners to discuss advances in the practice and theory of distributed computing. 👏 A huge shout out to our sponsors: Stellar Development Foundation | Mysten Labs | Neurolingo | Mediterranean Cloud Booking Services - MCBS | ICS-FORTH | University of Crete | HAR.S.H. (Hardware-Aware extReme-scale Similarity search) | Greece 2.0 | #NextGenerationEU

🎓 Last week, we were delighted to welcome high school students to the Department of Computer Science at the University of Crete! 🔬 During their visits, we had the opportunity to present our team's work and give them a glimpse into the future of technology. We are very proud to have participated in this wonderful exhibition. Η δράση υλοποιείται στο πλαίσιο του Εθνικού Σχεδίου Ανάκαμψης και Ανθεκτικότητας Ελλάδα 2.0 με τη χρηματοδότηση της Ευρωπαϊκής Ένωσης – NextGenerationEU The project is implemented under the National Recovery and Resilience Plan “Greece 2.0”, with funding from the European Union – NextGenerationEU

⭐️We are excited to share the following paper, “Evaluating and Generating Query Workloads for High-Dimensional Vector Similarity Search,” presented at ACM 2025. 📊The paper, written by Matteo Ceccarello, Alexandra Levchenko, Ioana Ileana, and Themis Palpanas, explores methods for efficiently generating and evaluating query workloads in high-dimensional vector similarity search. It introduces Hephaestus, a novel framework that enables researchers and practitioners to systematically create realistic workloads and benchmark similarity search systems, providing insights into performance, scalability, and accuracy. 📄 Paper: https://lnkd.in/dnFBtnqN Η δράση υλοποιείται στο πλαίσιο του Εθνικού Σχεδίου Ανάκαμψης και Ανθεκτικότητας Ελλάδα 2.0 με τη χρηματοδότηση της Ευρωπαϊκής Ένωσης – NextGenerationEU The project is implemented under the National Recovery and Resilience Plan “Greece 2.0”, with funding from the European Union – NextGenerationEU

⭐️We are thrilled to share our recent publication, Concurrent Balanced Augmented Trees, accepted at Symposium on Principles and Practice of Parallel Programming (PPoPP 2026).   📄The paper, written by Evan Wrench, Ajay Singh, Younghun Roh, Panagiota Fatourou, Siddhartha Jayanti, Eric Ruppert and Yuanhao Wei, explores a new way to create lock-free augmented balanced search trees.   📈This work presents the first lock-free augmented balanced search tree supporting generic augmentation functions. The experiments show that the suggested augmented balanced tree completes updates 2.2 to 30 times faster than the unbalanced augmented tree and outperforms unaugmented trees by several orders of magnitude on 120 threads. https://lnkd.in/deMqun3m Η δράση υλοποιείται στο πλαίσιο του Εθνικού Σχεδίου Ανάκαμψης και Ανθεκτικότητας Ελλάδα 2.0 με τη χρηματοδότηση της Ευρωπαϊκής Ένωσης – NextGenerationEU The project is implemented under the National Recovery and Resilience Plan “Greece 2.0”, with funding from the European Union – NextGenerationEU

⭐️We are thrilled to share our recent publication, Sharded Elimination and Combining for Highly-Efficient Concurrent Stacks, accepted at Symposium on Principles and Practice of Parallel Programming (PPoPP 2026).   📄The paper, written by Ajay Singh, Nikos Metaxakis and Panagiota Fatourou, explores the power of elimination for constructing a concurrent stack.   📈This work presents a new blocking linearizable stack implementation which utilizes sharding and fetch&increment to achieve significantly better performance than all existing concurrent stacks. https://lnkd.in/dTFMnvtU Η δράση υλοποιείται στο πλαίσιο του Εθνικού Σχεδίου Ανάκαμψης και Ανθεκτικότητας Ελλάδα 2.0 με τη χρηματοδότηση της Ευρωπαϊκής Ένωσης – NextGenerationEU The project is implemented under the National Recovery and Resilience Plan “Greece 2.0”, with funding from the European Union – NextGenerationEU

⭐️We are excited to share the following paper, “NILMFormer: Transformer-Based Approaches for Non-Intrusive Load Monitoring,” presented at ACM SIGKDD 2025 in Toronto. 📊The paper, written by Adrien Petralia, Philippe Charpentier, Youssef Kadhi, and Themis Palpanas, explores the use of transformer models for non-intrusive load monitoring (NILM). It introduces NILMFormer, a novel approach that leverages advanced sequence modeling to accurately disaggregate electrical consumption data, enabling more efficient and scalable energy usage analysis. 📄 Paper: https://lnkd.in/d8QnF3Ds Η δράση υλοποιείται στο πλαίσιο του Εθνικού Σχεδίου Ανάκαμψης και Ανθεκτικότητας Ελλάδα 2.0 με τη χρηματοδότηση της Ευρωπαϊκής Ένωσης – NextGenerationEU The project is implemented under the National Recovery and Resilience Plan “Greece 2.0”, with funding from the European Union – NextGenerationEU

⭐️We are pleased to announce that Themis Palpanas gave a plenary keynote on “Scalable Vector Analytics: A Story of Twists and Turns” at the Olympia Research Summit in Athens. 📊The talk highlighted the evolution of scalable vector analytics, sharing key challenges, lessons learned, and insights that have shaped modern approaches to high-dimensional data analysis in both research and real-world systems. Η δράση υλοποιείται στο πλαίσιο του Εθνικού Σχεδίου Ανάκαμψης και Ανθεκτικότητας Ελλάδα 2.0 με τη χρηματοδότηση της Ευρωπαϊκής Ένωσης – NextGenerationEU The project is implemented under the National Recovery and Resilience Plan “Greece 2.0”, with funding from the European Union – NextGenerationEU

⭐️We are proud to share that Themis Palpanas gave a talk on “Scalable Vector Analytics: A Story of Twists and Turns” at the Huawei Vision Forum in Paris. 📊The talk highlighted the evolution of scalable vector analytics, discussing key challenges, unexpected insights, and the techniques that have shaped modern approaches to high-dimensional data analysis in both research and real-world systems. Η δράση υλοποιείται στο πλαίσιο του Εθνικού Σχεδίου Ανάκαμψης και Ανθεκτικότητας Ελλάδα 2.0 με τη χρηματοδότηση της Ευρωπαϊκής Ένωσης – NextGenerationEU The project is implemented under the National Recovery and Resilience Plan “Greece 2.0”, with funding from the European Union – NextGenerationEU

⭐️We are pleased to announce that Professor Themis Palpanas gave a talk on high-dimensional similarity search at Archimedes Talks - Scalable Vector Analytics: A Story of Twists and Turns. 📊The talk explored the evolution of similarity search over the past fifty years, highlighting why this classic data management problem is more relevant and challenging than ever—from time-series management systems to modern vector databases. It presented state-of-the-art solutions spanning exact and approximate search, in-memory and on-disk methods, and techniques ranging from LSH and product quantization to k-NN graphs and optimized linear scans. A key insight was how recent advances in time-series similarity search now achieve state-of-the-art performance even for general high-dimensional vector data. We are proud to have contributed to this important discussion and to the ongoing exploration of open research directions. https://lnkd.in/dSvng7wr Η δράση υλοποιείται στο πλαίσιο του Εθνικού Σχεδίου Ανάκαμψης και Ανθεκτικότητας Ελλάδα 2.0 με τη χρηματοδότηση της Ευρωπαϊκής Ένωσης – NextGenerationEU The project is implemented under the National Recovery and Resilience Plan “Greece 2.0”, with funding from the European Union – NextGenerationEU

⭐️We are excited to share our recent paper, “TiVy: Time Series Visual Summary for Scalable Visualization,” which was presented at IEEE VIS 2025. 📊The paper, written by Gromit Yeuk-Yin Chan, Luis Gustavo Nonato, Themis Palpanas, Cláudio T. Silva, and Juliana Freire, introduces TiVy, a novel approach for scalable visualization of large collections of time series. TiVy provides compact visual summaries that preserve important patterns and trends, enabling analysts to efficiently explore and compare massive time-series datasets without sacrificing interpretability or performance. 📍 IEEE VIS 2025 🔗 Conference: https://lnkd.in/gCkCEkZN 📄 Paper: https://lnkd.in/dS5pg67q We’re excited to present this work and contribute to the discussion on scalable visual analytics for time-series data. Η δράση υλοποιείται στο πλαίσιο του Εθνικού Σχεδίου Ανάκαμψης και Ανθεκτικότητας Ελλάδα 2.0 με τη χρηματοδότηση της Ευρωπαϊκής Ένωσης – NextGenerationEU The project is implemented under the National Recovery and Resilience Plan “Greece 2.0”, with funding from the European Union – NextGenerationEU