Name
Abstract | ||
|---|---|---|
When data grows too large, we scale to larger systems, either by scaling out or up. It is understood that scale-out and scale-up have different complexities and bottlenecks but a thorough comparison of the two architectures is challenging because of the diversity of their programming interfaces, their significantly different system environments, and their sensitivity to workload specifics. In this paper, we propose a novel comparison framework based on MapReduce that accounts for the application, its requirements, and its input size by considering input, software, and hardware parameters. Part of this framework requires implementing scale-out properties on scale-up and we discuss the complex trade-offs, interactions, and dependencies of these properties for two specific case studies (word count and sort). This work lays the foundation for future work in quantifying design decisions and in building a system that automatically compares architectures and selects the best one. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1145/2534645.2534654 | DISCS@SC |
Field | DocType | Citations |
Data mining,Uncertainty quantification,Workload,Computer science,sort,Word count,Software,Data warehouse appliance,Scaling,Scalability | Conference | 8 |
PageRank | References | Authors |
0.90 | 11 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Michael Sevilla | 1 | 11 | 1.63 |
| Ike Nassi | 2 | 11 | 2.39 |
| Kleoni Ioannidou | 3 | 107 | 6.70 |
| Scott Brandt | 4 | 64 | 4.67 |
| Carlos Maltzahn | 5 | 1201 | 87.49 |