Abstract
Recent versions of Linux operating system employ O(1) scheduler to reduce the overhead in thread scheduling. However, O(1) scheduler does not take account of the affinity of threads which share the same memory address space. We propose a scheduling scheme to aggregate threads which share the same memory address space based on O(1) scheduler. Our scheduler can reduce context switches and exploit locality of reference. Our performance evaluation shows our scheduler can improve throughput of a chatroom server benchmark.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Molnar, I.: Ultra-scalable O(1) SMP and UP Scheduler, http://www.uwsg.iu.edu/hypermail/linux/kernel/0201.0/0810.html
Yamamura, S., Hirai, A., Sato, M., Yamamoto, M., Naruse, A., Kumon, K.: Speeding Up Kernel Scheduler by Reducing Cache Misses. In: Proceeding of the USENIX Annual Technical Conference 2002 FREENIX Track (2002)
Linux Benchmark Suite Homepage, http://lbs.sourceforge.net/
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Kondoh, T., Kusakabe, S. (2004). Context Aware Thread Aggregation in Linux. In: Liew, KM., Shen, H., See, S., Cai, W., Fan, P., Horiguchi, S. (eds) Parallel and Distributed Computing: Applications and Technologies. PDCAT 2004. Lecture Notes in Computer Science, vol 3320. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30501-9_160
Download citation
DOI: https://doi.org/10.1007/978-3-540-30501-9_160
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-24013-6
Online ISBN: 978-3-540-30501-9
eBook Packages: Computer ScienceComputer Science (R0)