Abstract | ||
---|---|---|
In heterogeneous CPU+GPU SoCs where a single DRAM is shared between both devices, concurrent memory accesses from both devices can lead to slowdowns due to memory interference. This prevents the deployment of real-time tasks, which need to be guaranteed to complete before a set deadline. However, freedom from interference can be guaranteed through software memory scheduling, but may come at a significant cost due to frequent CPU-GPU synchronizations. In this paper we provide a compile-time model to help developers make informed decisions on how to achieve freedom from interference at the lowest cost. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1145/3207719.3207735 | SCOPES '18: PROCEEDINGS OF THE 21ST INTERNATIONAL WORKSHOP ON SOFTWARE AND COMPILERS FOR EMBEDDED SYSTEMS |
Field | DocType | Citations |
Dram,Software deployment,Memory scheduling,Computer science,Parallel computing,Interference theory,Fault tolerance,Software,Interference (wave propagation),Dynamic priority scheduling,Embedded system | Conference | 0 |
PageRank | References | Authors |
0.34 | 5 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Björn Forsberg | 1 | 10 | 2.35 |
Luca Benini | 2 | 13116 | 1188.49 |
Andrea Marongiu | 3 | 337 | 39.19 |