ABSTRACT
Asymmetric multicore processors (AMP) consist of cores exposing the same instruction-set architecture (ISA) but varying in size, frequency, power consumption and performance. AMPs were shown to be more power efficient than conventional symmetric multicore processors, and it is therefore likely that future multicore systems will include cores of different types. AMPs derive their efficiency from core specialization: instruction streams can be assigned to run on the cores best suited to their demands for architectural resources. System efficiency is improved as a result. To perform effective matching of threads to cores, the thread scheduler must be asymmetry-aware; and while asymmetry-aware schedulers for operating systems are a well studied topic, asymmetry-awareness in hypervisors has not been addressed. A hypervisor must be asymmetry-aware to enable proper functioning of asymmetry-aware guest operating systems; otherwise they will be ineffective in virtual environments. Furthermore, a hypervisor must ensure that asymmetric cores are shared among multiple guests in a fair fashion or in accordance with their priorities.
This work for the first time implements simple changes to the hypervisor scheduler, required to make it asymmetry-aware, and evaluates the benefits and overheads of these asymmetry-aware mechanisms. Our evaluation was performed using an open source hypervisor Xen on a real multicore system where asymmetry was emulated via CPU frequency scaling. We compared the asymmetry-aware hypervisor to default Xen. Our results indicate that asymmetry support can be implemented with low overheads, and resulting performance improvements can be significant, reaching up to 36% in our experiments. Most performance improvements are derived from the fact that an asymmetry-aware hypervisor ensures that the fast cores do not go idle before slow cores and from the fact that it maps virtual cores to physical cores for asymmetry-aware guests according to the guest's expectations. Other benefits from asymmetry awareness are fairer sharing of computing resources among VMs and more stable execution times.
- M. Annavaram, E. Grochowski, and J. Shen. Mitigating Amdahl's Law through EPI Throttling. In Proc. of ISCA'05, pages 298--309, 2005. Google ScholarDigital Library
- S. Balakrishnan, R. Rajwar, M. Upton, and K. Lai. The Impact of Performance Asymmetry in Emerging Multicore Architectures. SIGARCH CAN, 33(2):506--517, 2005. Google ScholarDigital Library
- M. Becchi and P. Crowley. Dynamic Thread Assignment on Heterogeneous Multiprocessor Architectures. In Proc. of Computing Frontiers '06, pages 29--40, 2006. Google ScholarDigital Library
- C. Bienia, S. Kumar, J. P. Singh, and K. Li. The PARSEC Benchmark Suite: Characterization and Architectural Implications. In Proc. of PACT'08, October 2008. Google ScholarDigital Library
- L. Cherkasova, D. Gupta, and A. Vahdat. Comparison of the three CPU Schedulers in Xen. SIGMETRICS Perform. Eval. Rev., 35(2):42--51, 2007. Google ScholarDigital Library
- C. Evangelinos and C. N. Hill. Cloud Computing for Parallel Scientific HPC Applications: Feasibility of Running Coupled Atmosphere-Ocean Climate Models on Amazon's EC2. In Proc. Cloud Computing and Its Applications, 2008.Google Scholar
- M. D. Hill and M. R. Marty. Amdahl's Law in the Multicore Era. IEEE Computer, 41(7):33--38, 2008. Google ScholarDigital Library
- V. Kazempour, A. Fedorova, and P. Alagheband. Performance Implications of Cache Affinity on Multicore Processors. In Proc. of Euro-Par '08, pages 151--161, 2008. Google ScholarDigital Library
- J. Koomey. Estimating Total Power Consumption by Servers in the US and the World. Technical report, Lawrence Berkeley National Laboratory, February 2007.Google Scholar
- D. Koufaty, D. Reddy, and S. Hahn. Bias Scheduling in Heterogeneous Multicore Architectures. In Proceedings of the 5th ACM European Conference on Computer Systems (EuroSys), 2010. Google ScholarDigital Library
- R. Kumar, K. I. Farkas, and N. Jouppi et al. Single-ISA Heterogeneous Multi-Core Architectures: The Potential for Processor Power Reduction. In Proc. of MICRO 36, 2003. Google ScholarDigital Library
- R. Kumar, D. M. Tullsen, and P. Ranganathan et al. Single-ISA Heterogeneous Multi-Core Architectures for Multithreaded Workload Performance. In Proc. of ISCA '04. Google ScholarDigital Library
- T. Li, D. Baumberger, and D. A. Koufaty et al. Efficient Operating System Scheduling for Performance-Asymmetric Multi-Core Architectures. In Proc. of SC '07, pages 1--11. Google ScholarDigital Library
- T. Y. Morad, U. C. Weiser, A. Kolodny, M. Valero, and E. Ayguade. Performance, Power Efficiency and Scalability of Asymmetric Cluster Chip Multiprocessors. IEEE Comput. Archit. Lett., 5(1):4, 2006. Google ScholarDigital Library
- D. S. Nikolopoulos, G. Back, J. Tripathi, and M. Curtis-Maury. VTASOS: Holistic System Software Customization for Many Cores. pages 1--5, April 2008.Google Scholar
- J. Saez, A. Fedorova, M. Prieto, and H. Vegas. Operating System Support for Mitigating Software Scalability Bottlenecks on Asymmetric Multicore Processors. In Proceedings of the ACM International Conference on Computing Frontiers (CF), 2010. Google ScholarDigital Library
- J. Saez, M. Prieto, A. Fedorova, and S. Blagodurov. A Comprehensive Scheduler for Asymmetric Multicore Processors. In Proceedings of the 5th ACM European Conference on Computer Systems (EuroSys), 2010. Google ScholarDigital Library
- A. Shah and N. Krishnan. Optimization of Global Data Center Thermal Management Workload for Minimal Environmental and Economic Burden. Components and Packaging Technologies, IEEE Transactions on, 31(1):39--45, March 2008.Google Scholar
- D. Shelepov, J. C. Saez, S. Jeffery, A. Fedorova, Z. P. Huang, N. Perez, V. Kumar, and S. Blagodurov. HASS: A Scheduler for Heterogeneous Multicore Systems. ACM Operating System Review, 43(2), 2009. Google ScholarDigital Library
- T. Sterling and D. Stark. A High-Performance Computing Forecast: Partly Cloudy. Computing in Science and Engineering, 11(4):42--49, 2009. Google ScholarDigital Library
- Y. Xie and G. Loh. Dynamic Classification of Program Memory Behaviors in CMPs. In Proc. of CMP-MSI, 2008.Google Scholar
Index Terms
- AASH: an asymmetry-aware scheduler for hypervisors
Recommendations
AASH: an asymmetry-aware scheduler for hypervisors
VEE '10Asymmetric multicore processors (AMP) consist of cores exposing the same instruction-set architecture (ISA) but varying in size, frequency, power consumption and performance. AMPs were shown to be more power efficient than conventional symmetric ...
Virtualization-Based Operation Support Systems: Performance Evaluation and Systems Design
APNOMS '08: Proceedings of the 11th Asia-Pacific Symposium on Network Operations and Management: Challenges for Next Generation Network Operations and Service ManagementRecent advances in virtualization technologies have allowed telecommunications carriers to apply virtualization-based operation support systems. By separating logical view of computing resources from physical view, virtualization is expected to help ...
Virtualization-based separation of privilege: working with sensitive data in untrusted environment
VDTS '09: Proceedings of the 1st EuroSys Workshop on Virtualization Technology for Dependable SystemsContemporary commodity operating systems are too big and do not inspire trust in their security and reliability. Still they are used for processing sensitive data due to the vast amount of legacy software and good support for virtually all hardware ...
Comments