João Lourenço
Changing the world, one student at a time…
Computer Science Department, NOVA School of Science and Technology, NOVA University Lisbon, Quinta da Torre, P-2829-516 CAPARICA, Portugal — joao.lourenco [AT] fct [DOT] unl [DOT] pt (email)
Computer Science Department, NOVA School of Science and Technology, NOVA University Lisbon, Quinta da Torre, P-2829-516 CAPARICA, Portugal — joao.lourenco [AT] fct [DOT] unl [DOT] pt (email)
Funded by European Union | |
Euro-TM – Transactional Memories: Foundations, Algorithms, Tools, and Applications (EU Cost Action IC1001) |
Funded by National Science Foundation (FCT) | |
Synergy-VM: A Blueprint for the Next Generation Execution Environments |
|
RepComp - Replication of Components to Improve the Performance and Reliability of Multicore Systems |
Funded by European Union | |
Euro-TM – Transactional Memories: Foundations, Algorithms, Tools, and Applications (EU Cost Action IC1001) |
Funded by National Science Foundation (FCT) | |
Synergy-VM: A Blueprint for the Next Generation Execution Environments |
|
RepComp - Replication of Components to Improve the Performance and Reliability of Multicore Systems |
Principal Investigator: Dr. Nuno Preguiça
Coordinated by: CITI Funding Entity: FCT / MCTES Start Date: 2010-01-01 |
This project aims at investigating how the multiple cores in emerging multicore platforms can be used to improve the performance and reliability of applications. The idea is to use these cores to run replicated components that use alternate implementations (algorithms, data structures, etc.). Performance will be improved by using the best implementation for a particular operation, while reliability will be improved by using the redundancy available.
Multi-core processors have become widely available in the last couple of years. Although these processors are becoming increasingly powerful, with the integration of an increasing number of cores, it also makes it harder for applications to exploit this increasing power. Unlike before, that any application would automatically benefit from the increase of clock speed, with multicore CPUs, programs must include multiple concurrent threads of activity to take benefit from the multiple cores available. In this project we intend to explore the idea of diverse replication of software program components to achieve two goals. First, we expect to be able to improve the overall performance of an application by relying on the result obtained from the fastest replica in each operation (assuming that for different operations of a software component, the fastest component is not always the same). Second, we expect to be able to provide fault-tolerance to buggy implementations by using Btzantine fault-tolerance replication techniques.
Principal Investigator: Dr. João Lourenço Coordinated by: CITI Funding Entity: FCT / MCTES Start Date: 2011-03-01 Project Web Page: http://sites.fct.unl.pt/synergy-vm |
This project aims at investigating how to support application development in future generation clusters of multi-core computers, using Transactional Memory as the base programming paradigm.
This project aims at investigating how to support application development in clusters of multi-core machines. It proposes to provide a single system image to the programmer, based on the abstraction of the transactional memory, allowing the programs running on different cluster nodes to share a set of objects, and making use of the transactional paradigm to control the overall system consistency. To achieve good performance in such environment, the plan is to explore two levels of parallelism: coarser grain parallelism among computations on different nodes of the cluster; and finer grain parallelism for computations within each node. Coarser grain explore transaction boundaries for optimistic approach and validation, limiting the need for communication that was a performance-killer for most shared memory systems. Finer grain parallelism will exploit data locality and require moving highly dependent computations to a single node.
Principal Investigator: Dr. Paolo Romano
Coordinated by: INESC-ID Funding Entity: European Union Start Date: 2011-04-01 |
The Transactional Systems Team is partner in a EU funded 4 years long European Concerted Research Action (COST Action) named "Euro-TM — Transactional Memories: Foundations, Algorithms, Tools, and Applications". The action involves several academic and industrial partners in DK, FR, DE, GR, IE, IL, IT, PT, ES, SE, CH, UK.
Parallel programing used to be an area once confined to a few niches, such as scientific and high-performance computing applications. However, with the proliferation of multicore processors, and the emergence of new, inherently parallel and distributed deployment platforms, such as those provided by cloud computing, parallel programming has definitely become a mainstream concern. Transactional Memories (TMs) answer the need to find a better programming model for PP, capable of boosting developers' productivity and allowing ordinary programmers to unleash the power of parallel and distributed architectures avoiding the pitfalls of manual, lock based synchronization. It is therefore no surprise that TM has been subject to intense research in the last years. This action aims at consolidating European research on this important field, by coordinating the European research groups working on the development of complementary, interdisciplinary aspects of Transactional Memories, including theoretical foundations, algorithms, hardware and operating system support, language integration and development tools, and applications.