datalog is becoming increasingly popular as a standard tool for a variety of use cases. Modern datalog engines can achieve high performance by specializing datastructures for relational operations. For example, the D...
详细信息
ISBN:
(纸本)9781450383912
datalog is becoming increasingly popular as a standard tool for a variety of use cases. Modern datalog engines can achieve high performance by specializing datastructures for relational operations. For example, the datalog engine Souffle achieves high performance with a synthesizer that specializes datastructures for relations. However, the synthesizer cannot always be deployed, and a fast interpreter is required. This work introduces the design and implementation of the Souffle Tree Interpreter (STI). Key for the performance of the STI is the support for fast operations on relations. We obtain fast operations by de-specializing datastructures so that they can work in a virtual execution environment. Our new interpreter achieves a competitive performance slowdown between 1.32 and 5.67x when compared to synthesized code. If compile time overheads of the synthesizer are also considered, the interpreter can be 6.46x faster on average for the first run.
In this paper, we propose a hardware mechanism for embedded multi-core memory system called Pattern Aware Memory System (PAMS). The PAMS supports static and dynamic datastructures using descriptors and specialized me...
详细信息
ISBN:
(纸本)9781479959440
In this paper, we propose a hardware mechanism for embedded multi-core memory system called Pattern Aware Memory System (PAMS). The PAMS supports static and dynamic datastructures using descriptors and specialized memory and reduces area, cost, energy consumption and hit latency. When compared with a Baseline Memory System, the PAMS consumes between 3 and 9 times and 1.13 and 2.66 times less program memory for static and dynamic datastructures respectively. The benchmarking applications (having static and dynamic datastructures) results show that PAMS consumes 20% less hardware resources, 32% less on chip power and achieves a maximum speedup of 52x and 2.9x for static and dynamic datastructures respectively. The results show that the PAMS multi-core system transfers datastructures up to 4.65x faster than the MicroBlaze baseline system.
暂无评论