Share This Author
Dynamically Controlled Resource Allocation in SMT Processors
- F. Cazorla, A. Ramírez, M. Valero, E. Fernández
- Computer Science37th International Symposium on Microarchitecture…
- 4 December 2004
This paper proposes a novel resource allocation policy for SMT processors that directly monitors the usage of resources by each thread and guarantees that all threads get their fair share of the critical shared resources, avoiding monopolization.
Enabling preemptive multiprogramming on GPUs
- I. Tanasić, Isaac Gelado, Javier Cabezas, A. Ramírez, N. Navarro, M. Valero
- Computer ScienceACM/IEEE 41st International Symposium on Computer…
- 16 October 2014
This paper argues for preemptive multitasking and design two preemption mechanisms that can be used to implement GPU scheduling policies and extends the NVIDIA GK110 (Kepler) like GPU architecture to allow concurrent execution of GPU kernels from different user processes and implements a scheduling policy that dynamically distributes the GPU cores among concurrently running kernels, according to their priorities.
DiDi: Mitigating the Performance Impact of TLB Shootdowns Using a Shared TLB Directory
- C. Villavieja, Vasileios Karakostas, O. Unsal
- Computer ScienceInternational Conference on Parallel…
- 10 October 2011
This paper characterize the impact of TLB shoot downs on multiprocessor performance and scalability, and presents the design of a scalable TLB coherency mechanism that couples a shared TLB directory with load/store queue support for lightweight TLB invalidation, and thereby eliminates the need for costly IPIs.
Supercomputing with commodity CPUs: Are mobile SoCs ready for HPC?
- Nikola Rajovic, P. Carpenter, Isaac Gelado, Nikola Puzovic, A. Ramírez, M. Valero
- Computer ScienceSC - International Conference for High…
- 17 November 2013
The trend in mobile SoC performance is analyzed, comparing it with the similar trend in the 1990s, and a first answer as to whether mobile SoCs are ready for HPC is given.
Task Superscalar: An Out-of-Order Task Pipeline
- Yoav Etsion, F. Cabarcas, M. Valero
- Computer Science43rd Annual IEEE/ACM International Symposium on…
- 4 December 2010
The proposed task super scalar pipeline enables programmers to exploit many core systems effectively, while simultaneously simplifying their programming model.
Trace-driven simulation of multithreaded applications
- Alejandro Rico, A. Duran, F. Cabarcas, Yoav Etsion, A. Ramírez, M. Valero
- Computer Science(IEEE ISPASS) IEEE INTERNATIONAL SYMPOSIUM ON…
- 10 April 2011
A methodology to properly simulate multithreaded applications using trace-driven environments and creates several new research opportunities, including research on scheduling and other parallelism-management techniques for future architectures, and hardware support for programming models.
On the simulation of large-scale architectures using multiple application abstraction levels
These simulations show that a very high-level abstraction, which may be even faster than native execution, is useful for scalability studies on parallel applications; and that just simulating explicit memory transfers, they achieve accurate simulations for architectures using non-coherent scratchpad memories, with just a 25x slowdown compared to native execution.
Parallel Scalability of Video Decoders
- C. Meenderinck, A. Azevedo, B. Juurlink, M. Alvarez, A. Ramírez
- Computer ScienceJ. Signal Process. Syst.
- 1 November 2009
This work investigates the parallelism available in video decoders, an important application domain now and in the future, and proposes a new parallelization strategy, called Dynamic 3D-Wave, which allows certain MBs of consecutive frames to be decoded in parallel.
Tibidabo: Making the case for an ARM-based HPC system
The low-power architecture approach towards exascale computing
- Nikola Rajovic, Nikola Puzovic, L. Vilanova, C. Villavieja, A. Ramírez
- Computer ScienceScalA '11
- 14 November 2011
This work is exploring an alternative to current supercomputers that builds on the small energy-efficient mobile processors based on ARM Cortex-A9 to increase energy efficiency.