Name
Abstract | ||
|---|---|---|
Network processors usually consist of multiple heterogeneous processing and memory units connected by on-chip network, and the target applications generally need to process packets at full line rate from Gigabits to multiple 10Gigabits. Network processor based applications are real-time, resource-constrained and heterogeneous. We believe that component technology is a promising approach for complex embedded system development. The aim of this paper is to clarify the requirements of component technology for network processor based applications. We (1) analyze the software framework on IXP2400. There is a lack of formal specification and non-functional properties of component model in the existing software architecture, and few prediction mechanisms for global system behaviors are provided; (2) propose the component model, which contains performance facet that can be used to select component and predict the global performance; (3) discuss the frameworks basic services and available software QOS functionality in the frameworks. In the end, we present a prediction flow based on component to predict the global system performance, and the steps in the prediction flow are described using memory management as a case study. We show that component technology could improve predictability for network processor based applications. |
Year | DOI | Venue |
|---|---|---|
2005 | null | PROCEEDINGS OF THE SECOND IASTED INTERNATIONAL MULTI-CONFERENCE ON AUTOMATION, CONTROL, AND INFORMATION TECHNOLOGY - SOFTWARE ENGINEERING |
Keywords | Field | DocType |
component technology, performance prediction, network processor | Network processor,Computer architecture,Application-specific instruction-set processor,Computer science,Common Component Architecture,Information processor,Software system,Component-based software engineering | Conference |
Volume | Issue | ISSN |
null | null | null |
Citations | PageRank | References |
0 | 0.34 | 1 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Hong Xiao | 1 | 15 | 1.93 |
| Di Wu | 2 | 0 | 0.34 |
| Ling Zhang | 3 | 143 | 14.77 |