1 / 11

Scheduling Real-Time tasks on Symmetric Multiprocessor Platforms

Scheduling Real-Time tasks on Symmetric Multiprocessor Platforms. RETIS Lab. Real-Time Systems Laboratory. Research Area: Multiprocessor Systems. Marko Bertogna. Moore’s law effects. Pentium Tejas cancelled!. Power density (W/cm 2 ). Nuclear Reactor. P4. STOP. P3. Pentium P1. P2.

tyler
Télécharger la présentation

Scheduling Real-Time tasks on Symmetric Multiprocessor Platforms

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Scheduling Real-Time tasks on Symmetric Multiprocessor Platforms RETIS Lab Real-Time Systems Laboratory Research Area: Multiprocessor Systems Marko Bertogna

  2. Moore’s law effects Pentium Tejas cancelled! Power density (W/cm2) Nuclear Reactor P4 STOP P3 Pentium P1 P2 Hot-plate 286 486 8086 386 8085 8080 8008 4004 Year Clock speed limited to less than 4 GHz Leakage current intolerable @ 90nm

  3. Motivations • Improve computing performances at reasonable power consumption. • Multiprocessor-based architectures: • High-level computing: Intel’s Pentium D, Core 2 Duo, Itanium and Xeon; AMD’s Opteron, Quad FX and Athlon64 X2; etc. • Embedded market: TI’s OMAP, NXP’s Nexperia, STM’s Nomadik, ARM’s MPCore, Sony-IBM-Toshiba’s Cell, and many others. • How to program these devices?

  4. Problems • Extend OS capabilities to exploit parallel computation. • Adapt classical RT scheduling analysis to multi-core platforms. • Develop Real-Time Multi-Processor Operating Systems (MPRTOS). • Provide efficient tests to check schedulability of Multiprocessor systems. • Explore “global” scheduling solutions.

  5. Global scheduling on SMP CPU1 t1 CPU2 t2 t5 t4 t3 t2 t1 CPU3 t3 • Single system-wide queue: • Preemption and Migration. • Load balancing. • How to sort tasks in the ready queue? • Preemptive EDF is optimal only for uniprocessors. • Pfair is optimal for implicit deadline systems.

  6. Multiprocessor scheduling anomalies Utot 1 T • Scheduling problem is in general NP-hard. • Schedulability problem is as well NP-hard. • Dhall’s effect significantly degrades perfromances of classical scheduling algorithms. • Synchronous instant is not “critical”. • Only sufficient schedulability conditions. DEADLINE MISS

  7. Main results • Sufficient schedulability tests with (pseudo-) polynomial complexity for fixed and dynamic priority multiprocessor systems. • Scheduling algorithms that provide better performances on a SMP platform with bounded number of preemptions/migrations. • Strategies to derive timely characteristics of given task sets, like robustness, slack, sensitivity to dynamic load variations. • General analysis that can be extended for every global scheduling algorithm.

  8. Experimental results for EDF • 2 processors • Constrained deadlines • 1.000.000 task sets generated • Our test is constantly superior at all utilizations generated task sets our test Improvement over existing solutions Task set utilization

  9. Experimental results for FP • 2 processors • Constrained • deadlines • 1.000.000 • task sets • generated • Our test is • constantly • superior at all • utilizations generated task sets our test improvement Task set utilization

  10. Conclusions • Multiprocessor Real-Time systems are a promising field to explore. • Still few existing results far from tight conditions. • We contributed filling this gap. • Future work: • Find tighter schedulability tests. • Use our techniques to analyze the efficiency of other scheduling algorithms (EDZL, EDF-US, FP-DS, etc). • Take into account exclusive resources access. • Integrate into Resource Reservation framework.

  11. Marko BertognaPhD studentmarko@sssup.it RETIS Lab Real-Time Systems Laboratory Thank you

More Related