This is the authors' abstract. We don't add key points for this paper.
Recent advent of the asymmetric multi-core processors such as Cell Broadband Engine (Cell/BE) has popularized the use of heterogeneous architectures. A growing body of research is exploring the use of such architectures, especially in High-End Computing, for supporting scientific applications. However, prior research has focused on use of the available Cell/BE operating systems and runtime environments for supporting compute-intensive jobs. Data and I/O intensive workloads have largely been ignored in this domain. In this paper, we take the first steps in supporting I/O intensive workloads on the Cell/BE and deriving guidelines for optimizing the execution of I/O workloads on heterogeneous architectures. We explore various performance enhancing techniques for such workloads on an actual Cell/BE system. Among the techniques we explore, an asynchronous prefetching-based approach, which uses the PowerPC core of the Cell/BE for file prefetching and decentralized DMAs from the synergistic processing cores (SPE's), improves the performance for I/O workloads that include an encryption/decryption component by 22.2%, compared to I/O performed naïvely from the SPE's. Our evaluation shows promising results and lays the foundation for developing more efficient I/O support libraries for multi-core asymmetric architectures.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Rafique et al. (2008) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: