Abstract
The performance of many GPU-based systems depends heavily on the effective bandwidth for transferring data between the processors. For real-time systems, the importance of data transfer rates may be even higher due to non-deterministic transfer times that limit the ability to satisfy response time requirements. We present a new method that allows real-time applications to make efficient use of the communication infrastructure in multi-GPU systems, while retaining the necessary execution time predictability. Our method is based on a new application interface for executing batch operations composed of multiple command streams that can be executed in parallel. The new interface provides the run-time with information it needs to optimize the communication and to reduce the execution time. The method is compliant with common scheduling algorithms, such as EDF and RM, as it provides accurate offline execution time prediction for jobs using their definition and system characteristics.
Experiments with two multi-GPU systems show that our method achieves 7.9x shorter execution time than the bandwidth allocation method, and 39 % higher image resolution than the time division method, for realistic applications.
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
Zapata, O.U.P., Alvarez, P.M.: EDF and RM multiprocessor scheduling algorithms: Survey and performance evaluation. Queue, pp. 1–24 (2005)
Baruah, S., Goossens, J.: Handbook of Scheduling: Algorithms, Models, and Performance Analysis. Chapman Hall/CRC Press (2004)
NVIDIA Corporation, CUDA API Reference Manual, version 5.0 (2012)
Jeffay, K., Stanat, D., Martel, C.: On non-preemptive scheduling of period and sporadic tasks. In: Real-Time Systems Symposium, pp. 129–139 (1991)
Lehoczky, J.P., Sha, L.: Performance of real-time bus scheduling algorithms. ACM SIGMETRICS Performance Evaluation Review 14, 44–53 (1986)
Natale, M., Meschi, A.: Scheduling messages with earliest deadline techniques. Real-Time Systems (1993), 255–285 (2001)
Sinnen, O., Sousa, L.A., Member, S.: Communication contention in task scheduling. IEEE Transactions on Parallel and Distributed Systems 16(6), 503–515 (2005)
Balman, M.: Data transfer scheduling with advance reservation and provisioning. Ph.D. dissertation, Louisiana State University (2010)
Kato, S., Lakshmanan, K.: RGEM: A responsive GPGPU execution model for runtime engines. In: Real-Time Systems Symposium (RTSS), pp. 57–66 (November 2011)
Basaran, C., Kang, K.-D.: Supporting preemptive task executions and memory copies in GPGPUs. In: Euromicro Conference on Real-Time Systems, pp. 287–296 (July 2012)
Verner, U., Schuster, A., Silberstein, M., Mendelson, A.: Scheduling processing of real-time data streams on heterogeneous multi-GPU systems. In: International Systems and Storage Conference (SYSTOR), pp. 1–12 (2012)
Kato, S., Aumiller, J., Brandt, S.: Zero-copy I/O processing for low-latency GPU computing. In: International Conference on Cyber-Physical Systems (ICCPS 2013), pp. 170–178 (2013)
Augonnet, C., Clet-Ortega, J., Thibault, S., Namyst, R.: Data-aware task scheduling on multi-accelerator based platforms. In: International Conference on Parallel and Distributed Systems (ICPADS), pp. 291–298 (2010)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Verner, U., Mendelson, A., Schuster, A. (2014). Batch Method for Efficient Resource Sharing in Real-Time Multi-GPU Systems. In: Chatterjee, M., Cao, Jn., Kothapalli, K., Rajsbaum, S. (eds) Distributed Computing and Networking. ICDCN 2014. Lecture Notes in Computer Science, vol 8314. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-45249-9_23
Download citation
DOI: https://doi.org/10.1007/978-3-642-45249-9_23
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-45248-2
Online ISBN: 978-3-642-45249-9
eBook Packages: Computer ScienceComputer Science (R0)