Home arrow Project Description arrow Summary  

Main Menu
About us
Project Description
Quantitative Results
Research Lines
Research Results
Impact on Society
Press room
Contact us
Secure Login
Events Calendar
« < December 2019 > »
25 26 27 28 29 30 1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31 1 2 3 4 5


The High-performance, Reliable Architectures for Data Centers and Internet Servers (ref. CSD2006-00046) project aims at developing several techniques to improve the performance and reliability of current server architectures for data centers and Internet servers, for a given cost and power consumption budget. This project will develop research on processor, memory, and interconnect architectures, multiprocessor systems, system software, parallel file systems, dynamic load balancing, and techniques to efficiently support data acquisition and client accesses through wireless networks, as well as improving the quality of resource access through the Grid.

This project combines evolutionary and disruptive research with the implementation of a prototype. Among the disruptive research activities, we will develop techniques to commit instructions out of order without checkpointing which will drastically reduce the negative impact of increasing memory latencies, a cacheless memory organization that will reduce average memory access time, an interconnection network architecture with an order of magnitude lower packet latency than existing ones, and will study techniques toward automatic extraction of non-speculative thread-level parallelism in hardware. This project will also deliver new results along the research lines traditionally followed by our group, including research on processor microarchitecture, reduction of processor power consumption, chip multiprocessors, cache coherence protocols, switch architecture, congestion management, routing, fault tolerant networks, reduction of network power consumption, support for quality of service (both in the interconnection network itself and in server access through the Grid), network reconfiguration, techniques to efficiently support wireless access to servers, parallel file systems, system software for parallel computers, and load balancing techniques.

Finally, some of these techniques will be implemented, tuned, and evaluated on a testbed to show the feasibility of the proposed approaches.