introductory seminar on research cis5935 fall 2009
DESCRIPTION
Introductory Seminar on Research CIS5935 Fall 2009. Ted Baker. Outline. Introduction to myself My past & current research areas Technical background for the remainder A recent paper, on Sustainable MP Scheduling What is it? Why do I want it? How do I get it? - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/1.jpg)
Introductory Seminar on Research
CIS5935 Fall 2009
Ted Baker
![Page 2: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/2.jpg)
Outline
• Introduction to myself– My past & current research areas
• Technical background for the remainder• A recent paper, on Sustainable MP Scheduling
– What is it?– Why do I want it?– How do I get it?– Sustainability of specific algorithms & tests
![Page 3: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/3.jpg)
My past research
• Relative computability– Relativizations of the P=NP? question (1975-1979)
• Algorithms– N-dim pattern matching (1978)– extended LR parsing (1981)
• Compilers & PL implementation– Ada compiler and runtime systems (1979-1998)
• Real-time runtime systems, multi-threading– FSU Pthreads & other RT OS projects (1985-1998)
• Real-time scheduling & synch.– Stack Resource Protocol (1991)– Deadline Sporadic Server (1995)
• RT Software standards– POSIX, Ada (1987-1999)
![Page 4: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/4.jpg)
Recent/current research
• Multiprocessor real-time scheduling (1998-…)
– how to guarantee deadlines for task systems scheduled on multiprocessors?
with M. Cirinei & M. Bertogna (Pisa), N. Fisher & S. Baruah (UNC)
• Real-time device drivers (2006-…)
– how to support schedulability analysis with an operating system?
– how to get predictable I/O response times?with A. Wang & Mark Stanovich (FSU)
![Page 5: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/5.jpg)
Background
• Real-time scheduling goals & problems• Workload models• Scheduling algorithms & tests
![Page 6: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/6.jpg)
Real-time scheduling
Context = embedded systems, controllersProblems =• How to schedule a computational workload to
meet timing constraints? (deadlines)• How to do this while meeting other constraints
and goals? (energy, fault tolerance, etc.)• How to guarantee constraints will always be met,
in the face of workload variations
![Page 7: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/7.jpg)
Typical quality questions
• Is a given system schedulable to meet deadlines?
• Is a given system schedulable by a given algorithm?
• If deadlines are missed, by how much?• How good is a given algorithm at finding a
feasible schedule?• How sensitive is a schedulability
guarantee to variations in parameters?
![Page 8: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/8.jpg)
Workload models
• job = schedulable unit of computation, with– arrival time– worst-case execution time (WCET)– deadline
• task = sequence of jobs• task system = set of tasks• independent tasks:
can be scheduled without consideration of interactions, precedence, coordination, etc.
![Page 9: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/9.jpg)
Example: sporadic task i• Ti = minimum inter-arrival time
• Ci = worst-case execution time
• Di = relative deadline
job released job completes deadline next release
iC
?iDiD
iT
scheduling window
![Page 10: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/10.jpg)
Multiprocessor scheduling
• m identical processors (vs. uniform/hetero.)
• shared memory (vs. distributed)• Kinds of algorithms:
– preemptive vs. non-preemptive– on-line vs. off-line– global vs. partitioned
![Page 11: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/11.jpg)
Optimality
• There is no optimal on-line global scheduling algorithm for sporadic tasks [Fisher, 2007]
• But there are some good algorithms
![Page 12: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/12.jpg)
Some good scheduling algorithms
• Fixed task-priority– global, or partitioned, hybrids, …
• EDF– global or partitioned, hybrids, …
• EDZL…
![Page 13: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/13.jpg)
Example: Global EDF
• m identical processors• global =
– jobs can run on any processor– can be preempted, and migrated
• EDF = “earliest deadline first”– earlier deadline implies higher priority– the top m jobs will execute– no processor is idle unless there are
fewer than m jobs ready to run
![Page 14: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/14.jpg)
Schedulability testing
Global-EDF schedulability for sporadic task systems can be decided by brute-force state-space enumeration (in exponential time) [Baker, OPODIS 2007]
but we don’t have any practical algorithm.
We do have several practical sufficient conditions.
![Page 15: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/15.jpg)
Sufficient Conditions for Global EDF
• Varying degrees of complexity and accuracy
• Examples: • Goossens, Funk, Baruah: density test (2003)• Baker: analysis of -busy interval (2003)• Bertogna, Cirinei: iterative slack time
estimation (2007)
• Difficult to compare quality, except by experimentation
• All tests are very conservative
![Page 16: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/16.jpg)
Example: Density Test
)()1()( max mmsum
n
iisum
1
i
n
i
1max max
),min( ii
ii TD
C
where
Sporadic task system is schedulable on m unit-capacity processors if
![Page 17: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/17.jpg)
My recent researchon Sustainable Schedulabilty
done with Sanjoy Baruah (UNC)presented by T. Baker at ECRTS
July 2009Dublin
![Page 18: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/18.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 18
General schedulability testing
Will a will workload W meet all deadlines ifscheduled by algorithm A on platform P?
exact test : “yes” ⇔ schedulablesufficient test : “yes” ⇒schedulable
Goal: validate timing propertieson which correctness, reliability, safety depend.
![Page 19: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/19.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 19
Potential weakness• Real systems differ from models
– clock and timer inaccuracies– blocking/non-preemption– execution time variation
• Differences must not invalidate test• Solution:
– pessimistic assumptions
![Page 20: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/20.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 20
What is pessimistic?
• Should cover full range of possibilities.• How to validate over such a range?• People tend to assume system behavior is
monotonic, and rely on boundary testing.• This can be dangerous.• For example, does WCET of individual jobs
give worst-case behavior of the entire system?
![Page 21: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/21.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 21
Goals
• Our notions of “pessimistic” should be consistent with the analysis that underlies the schedulability test.
• The test should be monotonic with respect to any workload parameters that are considered to be bounds on actual values.
![Page 22: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/22.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 22
Predictability(Ha & Liu 1994)
• For arbitrary job sets on multiprocessors, if scheduling algorithm is work-preserving, preemptive, global (migration), with fixed job priorities:– Job completions times are monotonically
related to job execution times.– It is safe to consider only upper bounds
for job execution times in schedulability tests.
• Not true for non-preemptive scheduling
![Page 23: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/23.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 23
Sustainabilty(Baruah & Burns 2006)
• Extends idea to a wider range of relaxations of workload parameters:– shorter execution times– longer periods– less release-time jitter– later deadlines
• Relaxation should preserve schedulability• They considered single-processor systems• We consider multiprocessors
![Page 24: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/24.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 24
Why to relax?
• Safety– Run-time behavior may fall short of
pessimistic model parameters.• Modular design
– Start with pessimistic parameters, relax as design evolves.
• Efficiency– Revise estimates at job arrival, schedule
more precisely.
![Page 25: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/25.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 25
How?
• Actual/nominal?
– actual max. job exec. time < Ci, or
– value Ci used by scheduler is reduced
• Per job/task?
– Ci reduced differently for each job, or
– Ci reduced for all jobs of task
• Use in scheduling decisions, or not?These may or may not make a difference in
sustainability, depending on context.
![Page 26: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/26.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 26
Getting specific
• Sustainability depends on the platform and workload model as well as the scheduling algorithm.
• Platform : SMP• Workload model : sporadic task set• Scheduling algorithms : global FTP, EDF,
EDZL
![Page 27: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/27.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 27
SMP Platform
• m identical processors• bounded cost for migration, context switch
![Page 28: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/28.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 28
Sporadic task workload
• fixed set of tasks• each generates a sequence of jobs• each job of a task has parameters:
– exec. time• Ci is an upper bound
– inter-arrival time• Ti is a lower bound
– deadline, relative to arrival time, Di
![Page 29: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/29.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 29
Where is pessimism?
• Ti : lower bound
• Ci : upper bound
• Di : lower bound?
So sustainability means allowing for the following relaxations:
• Longer inter-arrival times• Shorter execution times• Longer deadlines
![Page 30: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/30.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 30
Global Scheduling
• Fixed Task Priority (FTP): m tasks with highest task priority allowed to execute
• EDF: m jobs with earliest deadlines allowed to execute
• EDZL: same, except jobs with zero laxity preemptlaxity = remaining-time-to-deadline – remaining-WCET
Common property: Limit of one (or 2, for EDZL) preemption per job allows us to bundle worst-case preemption/migration overhead into WCET of jobs
![Page 31: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/31.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 31
Longer inter-arrival times
• Sustainability is built into the sporadic workload model– per-job or per-task– actual, nominal too for algorithms that
do not make use of Ti in scheduling decisions
• So global FTP, EDF, EDZL are sustainable w.r.t. longer inter-arrival times
![Page 32: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/32.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 32
Shorter execution times
• Depends on scheduling algorithm• Studied by Ha & Liu (1994) • Non-preemptive scheduling not sustainable• Preemptive work-conserving fixed-job-
priority SMP scheduling with migration is sustainable– FTP and EDF are sustainable
• Proved for job sets, covers sporadic tasks– actual/nominal, per job/task
![Page 33: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/33.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 33
EDZL & shorter exec. time
• Difference from FTP & EDF: job priority can change (breaks Ha & Liu assumption)
• EDZL is sustainable
– per-job/task– nominal and actual, or just actual
(Piao, Han, Kim et al. 2006; new proof here)
![Page 34: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/34.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 34
Longer deadlines?
• FTP is sustainable, for sporadic tasks• EDF & EDZL sustainability open for
sporadic tasks(not sustainable for job sets*)– exact test (if one exists) might be risky– or scheduler should ignore deadline
changes– sustainable tests are of interest*Apparently also not sustainable for sporadic task sets,
with respect to relaxation of a single job deadline (see appendix).However, it might be sustainable w.r.t. uniform relaxation of deadlineover all jobs of a task.
![Page 35: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/35.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 35
Sustainable schedulability test
• If the test says a task set is schedulable, then it remains schedulable if we reduce pessimism in our assumptions, e.g., – decrease task execution times– increase task periods– reduce release-time jitter– lengthen job deadlines
![Page 36: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/36.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 36
Relationships
• Algorithm sustainable ⇒ test sustainable• Algorithm not sustainable ⇒
exact test not sustainable
So an inexact test might be better, if it is sustainable and the algorithm is not.
This might be true for global EDF.Do we know any sustainable tests?
![Page 37: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/37.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 37
Self-sustainable test(new)
• If the test says a task set is schedulable, then the test still says it is schedulable if we reduce pessimism in our assumptions, e.g.– decrease task execution times– increase task periods– reduce release-time jitter– lengthen job deadlinesDo we know any self-sustainable tests for global EDF?
![Page 38: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/38.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 38
BCL Test
min (Dk Ck ),Ii,k ik
m(Dk Ck )
or
Sporadic task system is global EDF schedulable on m unit-capacity processors if for each
min (Dk Ck ),Ii,k ik
m(Dk Ck )
(Bertogna, Cirinei, Lipari ECRTS 2003)
and there is some i≠k for which
k
Ii,k Dk Ck
N i,k (Dk Di) /Ti 1
Ii,k N i,kCi min(Ci,max(0,Dk N i,kTi))
![Page 39: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/39.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 39
• We don’t know.• We have proven BCL test is not self-
sustainable with respect to nominal deadline relaxation.
Is it Sustainable?
![Page 40: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/40.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 40
A load-based test
Sporadic task system is global-EDF schedulable on m unit-capacity processors if
where
)()1()( max LOAD
)()1( max mm
(Baruah & Baker, ECRTS 2008)
DBF( i, t)max(0,(t DiTi
1)Ci)
LOAD()maxt0
DBF( i,t) / ti1
n
![Page 41: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/41.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 41
Is it sustainable?
• We don’t know, but it is not self-sustainable.
• Is there a better test, that is sustainable?• Yes, this is self-sustainable:
)()2(1
)()1(max)(
max
max
LOAD
![Page 42: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/42.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 42
Why one is sustainable and the other not
() ( ( ) 1)max ( )
Load
( ) 1 ( ( ) 2)max ( )
not schedulable
schedulable
![Page 43: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/43.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 43
Summary
• Sustainability depends on workload modelas well as platform and scheduling algorithm
• For sporadic tasks and SMP:– Global FTP is fully sustainable– Global EDF & EDZL are (only?) exec. time and inter-
arrival time sustainable
• Self-sustainability is a strong form of sustainability for tests– Not all tests are deadline self-sustainable– But at least one is
![Page 44: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/44.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 44
Suggestion
• Papers on scheduling algorithms, and schedulability tests, should routinely address sustainability.
![Page 45: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/45.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 45
The End
questions?
![Page 46: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/46.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 46
Appendix 1
A tentative proof that global EDF is not sustainable for sporadic task
systems w.r.t. relaxation of individual job deadlines
![Page 47: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/47.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 47
Example
Task WCET (Ci) Deadline (Di) Period (Ti)
A 2 4 8B 2 4 8C 4 8 8D 8 13 13
B
D
A
C
0 62 4 148 10
m = 2 processors
1816 2012
A
B
C
![Page 48: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/48.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 48
Claim
• This task appears to be global EDF schedulable.
• This checks out for integer time values up to a scale factor of 3, using brute-force test.
• But it still needs a careful proof.• If this task set is schedulable, it shows that
global EDF is not sustainable for sporadic task systems w.r.t. single-job deadline relaxation.
![Page 49: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/49.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 49
If B’s first deadline is relaxed?
B
D
A
0 62 4 148 10 1816 2012
A
BC
D can be blocked for 6 time units, missing its deadline.This does not happen
if both of B’s deadlines are relaxed equally.
D
![Page 50: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/50.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 50
If B is just released later?
B
D
A
0 62 4 148 10 1816 2012
A B
C
D is blocked for only 4 time units,since the second B has later deadline than D.
![Page 51: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/51.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 51
Appendix 2
Another self-sustainable test for global EDF on sporadic task sets.
![Page 52: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/52.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 52
Density Test for Global EDF*
)()1()( max mmsum
n
iisum
1
i
n
i
1max max
),min( ii
ii TD
C
where
Sporadic task system is schedulable on m unit-capacity processors if
(Goossens, Funk, Baruah RTS 2003)
7/3/2009 52ECRTS 2009 (Dublin) - Ted Baker
This is another test for global EDF that is fully self-sustainable.
![Page 53: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/53.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 53
Appendix 3
Discussion of Sustainability & Workload Models
![Page 54: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/54.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 54
Job Workload models
• A job has an arrival time, deadline, and execution time.
• The execution time may depend on how the job’s execution is scheduled, but there is an upper bound.
• A workload model specifies a collection of possible sets of jobs.
• It is schedulable by an algorithm if every set of jobs that conforms to the model is scheduled to meet all deadlines.
![Page 55: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/55.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 55
Kinds of sustainability
1. With respect to differences between actual run-time behavior and model.– e.g., actual execution time is less than
nominal WCET.– a property of the model
2. With respect to changes in the model.– e.g., nominal WCET is reduced– a property of the model and scheduling
algorithm
![Page 56: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/56.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 56
Sustainability of Model
• Sustainability can be built into a workload model, e.g., specify for each task
– maximum job execution time, Ci
– minimum inter-arrival time, Ti
– shortest relative deadline, Di
• Otherwise, it may be imposed by the scheduling algorithm.
![Page 57: Introductory Seminar on Research CIS5935 Fall 2009](https://reader036.vdocuments.site/reader036/viewer/2022062410/56815a05550346895dc75162/html5/thumbnails/57.jpg)
7/3/2009 ECRTS 2009 (Dublin) - Ted Baker 57
Sustainable scheduling algorithm
• If all job sets satisfying a model are schedulable, then sets of jobs that falls short of the model’s assumptions are also schedulable, e.g.,– shorter execution times– longer periods– less release-time jitter– later deadlines