Run-Time Scalable Systolic Coprocessors for Flexible Multimedia SoPCs

Abstract

Abstract-Multimedia Systems on Chip have high computational requirements, as well as significant flexibility demands. Flexibility can be related with the reusability of the cores in charge of the execution of computation-intensive tasks, but also with the run-time adaptation of these cores to the execution of time-variable tasks, or to changing system conditions. Among the run-time flexibility requirements of hardware IPs, functional scalability has been identified as an interesting feature. The proposal in this paper is to take advantage of the regularity and the high-processing capability of systolic arrays, to develop runtime functional scalable cores, making use of spatial scalability, by means of replicating and relocating basic processing elements of the array. The relocation process is performed using the dynamic-reconfiguration possibilities offered by commercial FPGAs. In this paper, an architectural template is proposed to develop systolic scalable coprocessors following this approach, together with its corresponding software drivers that may be executed within an embedded processor. In addition, a design flow is proposed to adapt the architectural template to different problems, together with some examples of scalable cores created following this design. This solution provides better results, regarding the reconfiguration time and the memory necessity overhead, compared with other dynamically scalable solutions.

A fast Reconfigurable 2D HW core architecture on FPGAs for evolvable Self-Adaptive Systems

“…Future work will address an improved implementation to optimize the area occupancy of each PE and the delay introduced by the Bus Macros, further improving the maximum operating frequency, even eliminating them as in [7]. In addition, the number of evaluations needed to obtain a working solution will try to be reduced, so a different EA, such as a Genetic Algorithm will be tested.…”

A Novel FPGA-based Evolvable Hardware System Based on Multiple Processing Arrays

“…For instance, scalability can be static or dynamic, depending on whether the number of arrays is defined at design time or at run-time. In this work, scalability for a variable number of arrays was implemented using the same design principles as in [15], where generic dynamically scalable processing cores are presented. Main proposal of this work is that a change of dimensions of the architecture is tackled with a proportional change of the footprint of the core, leading to scalable footprints depending on performance or functional requirements.…”

A novel scalable Deblocking Filter architecture for H.264/AVC and SVC video codecs

“…This technique has been applied on a novel coarse grain architecture based on a systolic array structure, adapting the original proposal provided in [5]. This means that different FUs work in parallel, like a multiprocessor system, but with the performance advantage of using specific elements instead of general purpose microprocessor cores.…”

“…Regarding to this, [5], [6], and [7] use different memory accessing techniques in order to satisfy real-time constraints. With the same purpose, [8] and [9] explore internal data parallelism.…”