Publications
Detailed Information
MLB: A Memory-aware Load Balancing Method for Mitigating Memory Contention : 메모리 경쟁을 완화 시킬 수 있는 메모리 인지 로드 밸런스 기법
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 엄현상 | - |
dc.contributor.author | 서동유 | - |
dc.date.accessioned | 2017-07-14T02:54:07Z | - |
dc.date.available | 2017-07-14T02:54:07Z | - |
dc.date.issued | 2014-02 | - |
dc.identifier.other | 000000016925 | - |
dc.identifier.uri | https://hdl.handle.net/10371/123032 | - |
dc.description | 학위논문 (석사)-- 서울대학교 대학원 : 전기·컴퓨터공학부, 2014. 2. 엄현상. | - |
dc.description.abstract | Most of the current CPUs have not single cores, but multicores integrated in
the Symmetric MultiProcessing (SMP) architecture, which share the resources such as Last Level Cache (LLC) and Integrated Memory Controller (IMC). On the SMP platforms, the contention for the resources may lead to huge performance degradation. To mitigate the contention, various methods were developed | - |
dc.description.abstract | most of these methods focus on finding which tasks share the same
resource assuming that a task is the sole owner of a CPU core. However, task arrival patterns and the demand for resources in the current server environment are highly dynamic | - |
dc.description.abstract | hence, tasks, the number of which is larger than
that of CPU cores, can be executed simultaneously on a CPU. In order to mitigate contention for memory subsystems in such multitasking cases, dealing with the dynamicity of resource demand, we have devised a Memory-aware Load Balancing (MLB) method. MLB dynamically recogrnizes contention by using simple contention models and performs inter-core task migration to mitigate the contention. We have evaluated MLB on an Intel i7-2600 (desktop-level CPU) and a Xeon E5-2690 (server-level CPU), and found that our approach can be effectively taken in an adaptive manner, leading to noticeable performance improvements of memory intensive tasks (about 15% in best case) on the different CPU platforms. Also, MLB can achieve performance improvements in CPU-GPU communication in discrete GPU systems. | - |
dc.description.tableofcontents | Abstract i
Contents ii List of Figures iv List of Tables vi 0.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0.2 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 0.3 Motivations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 0.4 Memory Contention Modeling . . . . . . . . . . . . . . . . . . . . 12 0.4.1 Memory contention level . . . . . . . . . . . . . . . . . . . 12 0.4.2 The correlation between memory contention level and performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 0.5 The Design and Implementation of Memory-aware Load Balancing( MLB) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 0.5.1 Target runqueue and non target runqueue lists . . . . . . . 19 0.5.2 Predicted number . . . . . . . . . . . . . . . . . . . . . . . 20 0.5.3 Memory-aware Load Balacing algorithm . . . . . . . . . . . 22 ii 0.6 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 0.6.1 Dynamicity . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 0.6.2 Mitigation . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 0.6.3 Performance . . . . . . . . . . . . . . . . . . . . . . . . . . 32 0.6.4 Additional benefit . . . . . . . . . . . . . . . . . . . . . . . 33 0.6.5 Overhead . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 0.7 Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 0.8 Conclusions and Future Work . . . . . . . . . . . . . . . . . . . . 40 Bibliography 41 요약 46 Acknowledgements 47 | - |
dc.format | application/pdf | - |
dc.format.extent | 785445 bytes | - |
dc.format.medium | application/pdf | - |
dc.language.iso | en | - |
dc.publisher | 서울대학교 대학원 | - |
dc.subject | Multicore processors | - |
dc.subject | shared resource contention | - |
dc.subject | load balancing | - |
dc.subject | dynamicity | - |
dc.subject | SMP platform | - |
dc.subject.ddc | 621 | - |
dc.title | MLB: A Memory-aware Load Balancing Method for Mitigating Memory Contention | - |
dc.title.alternative | 메모리 경쟁을 완화 시킬 수 있는 메모리 인지 로드 밸런스 기법 | - |
dc.type | Thesis | - |
dc.description.degree | Master | - |
dc.citation.pages | 56 | - |
dc.contributor.affiliation | 공과대학 전기·컴퓨터공학부 | - |
dc.date.awarded | 2014-02 | - |
- Appears in Collections:
- Files in This Item:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.