One of the most critical aspects of integrating loosely-coupled accelerators in heterogeneous SoC architectures is orchestrating their interactions with the memory hierarchy, especially in terms of navigating the various cache-coherence options: from
accelerators accessing off-chip memory directly, bypassing the cache hierarchy, to accelerators having their own private cache. By running real-size applications on FPGA-based prototypes of many-accelerator multi-core SoCs, we show that the best cache-coherence mode for a given accelerator varies at runtime, depending on the accelerators characteristics, the workload size, and the overall SoC status. Cohmeleon applies reinforcement learning to select the best coherence mode for each accelerator dynamically at runtime, as opposed to statically at design time. It makes these selections adaptively, by continuously observing the system and measuring its performance. Cohmeleon is accelerator-agnostic, architecture-independent, and it requires minimal hardware support. Cohmeleon is also transparent to application programmers and has a negligible software overhead. FPGA-based experiments show that our runtime approach offers, on average, a 38% speedup with a 66% reduction of off-chip memory accesses compared to state-of-the-art design-time approaches. Moreover, it can match runtime solutions that are manually tuned for the target architecture.
A {it $k$-uniform hypergraph} $mathcal{H}=(V, E)$ consists of a set $V$ of vertices and a set $E$ of hyperedges ($k$-hyperedges), which is a family of $k$-subsets of $V$. A {it forbidden $k$-homogeneous (or forbidden $k$-hypergraph)} access structure
$mathcal{A}$ is represented by a $k$-uniform hypergraph $mathcal{H}=(V, E)$ and has the following property: a set of vertices (participants) can reconstruct the secret value from their shares in the secret sharing scheme if they are connected by a $k$-hyperedge or their size is at least $k+1$. A forbidden $k$-homogeneous access structure has been studied by many authors under the terminology of $k$-uniform access structures. In this paper, we provide efficient constructions on the total share size of linear secret sharing schemes for sparse and dense $k$-uniform access structures for a constant $k$ using the hypergraph decomposition technique and the monotone span programs.
High-Level Synthesis (HLS) frameworks allow to easily specify a large number of variants of the same hardware design by only acting on optimization directives. Nonetheless, the hardware synthesis of implementations for all possible combinations of di
rective values is impractical even for simple designs. Addressing this shortcoming, many HLS Design Space Exploration (DSE) strategies have been proposed to devise directive settings leading to high-quality implementations while limiting the number of synthesis runs. All these works require considerable efforts to validate the proposed strategies and/or to build the knowledge base employed to tune abstract models, as both tasks mandate the syntheses of large collections of implementations. Currently, such data gathering is performed ad-hoc, a) leading to a lack of standardization, hampering comparisons between DSE alternatives, and b) posing a very high burden to researchers willing to develop novel DSE strategies. Against this backdrop, we here introduce DB4HLS, a database of exhaustive HLS explorations comprising more than 100000 design points collected over 4 years of synthesis time. The open structure of DB4HLS allows the incremental integration of new DSEs, which can be easily defined with a dedicated domain-specific language. We think that of our database, available at https://www.db4hls.inf.usi.ch/, will be a valuable tool for the research community investigating automated strategies for the optimization of HLS-based hardware designs.
