Self-tuning Reactive Distributed Trees for Counting and Balancing

1
Self-tuning Reactive Distributed Trees for
Counting and Balancing

• Phuong Hoai Ha
• Marina Papatriantafilou
• Philippas Tsigas

OPODIS 04, Grenoble, France Dec. 15 17, 2004
2
Schedule

• Introduction
• Coordinating Objects
• Reactive Shared Objects
• Self-tuning Reactive Trees
• Algorithms
• Evaluation
• Conclusions

3
What are coordinating objects?
Coordination
?
?
?
high collision!
high load!
?
?
?

• Data structures that evenly distribute processes
  into small groups, which each accesses different
  shared objects in a coordinated manner.

4
Example Diffracting trees
B2
A0
B1
C6
B1
A
B,
F, E, D, C,
B3

• A highly distributed data structure
• Small groups of processes locally access shared
  data in a global coordinated manner.
• A disadvantage
• Optimal only for a small range of contention
  levels

5
Schedule

• Introduction
• Coordinating Objects
• Reactive Shared Objects
• Self-tuning Reactive Trees
• Algorithms
• Evaluation
• Conclusions

6
Why are reactive objects useful?

• Performance of concurrent objects heavily rely on
  their surrounding environment.
• Challenges
• In multiprocessor systems, processors
  surrounding environment changes too fast compared
  with their reactions.
• Multiprogramming systems are unpredictable
• Interference among processes ? traffic on the
  bus/network, contention on memory modules, load
  on processors etc. are unpredictable.

?
?
?
?
?
?
?
?
?
7
Example Reactive diffracting trees
A0
B1
B1
A
B,
F, E, D, C,
B3

• Each counter locally react according to its
  current load.
• Disadvantages
• Require experimentally tuned parameters for each
  system and each application
• Inefficient reactive scheme
• Shrink/expand one level in one adjustment step
• Costly adjustment

8
Reactive policy
9
Schedule

• Introduction
• Coordinating Objects
• Reactive Shared Objects
• Self-tuning Reactive Trees
• Algorithms
• Evaluation
• Conclusions

10
Self-tuning trees

• Advantages
• No need of any experimentally tuned parameters
• Efficient reactive scheme
• Ability to shrink and expand many levels at one
  time
• Reasonable costs for adjustments.

11
Reactive policy

• Problem balancing the trade-off between two key
  measures, the contention level and the depth of
  the tree.

?
?
?
?
?
?
?
?
?
?
?
12
When to expand/shrink ?
Rules for expansion (cf. threat-based algorithm)

• Advantages
• Reduce contention keep traversal short
• Favor to expand leaves with high contention.
• Similar for shrinkage

13
Expanding a leaf to a subtree
IN
12
2
8
2 pending processors
10
6
14
Shrinking a subtree to a leaf

• Elect2Shrink check whether half the number of
  leaves in the subtree vote for the leaf
• Shrink

IN
8
12
3/4 ? shrink
4
6
5 active processors
2
15
Results Full contention benchmark
16
Results Index distribution benchmark
17
Results Surge load benchmark
18
Schedule

• Introduction
• Coordinating Objects
• Reactive Shared Objects
• Self-tuning Reactive Trees
• Algorithms
• Evaluation
• Conclusions

19
Conclusions

• We have presented a new data structure that
• Distributes a set of processors to many smaller
  groups accessing disjoint critical sections in a
  global coordinated manner.
• Collects information about the contention at the
  leaves and then efficiently adjusts itself to
  attain optimal performance without using any
  experimental parameters.
• Methodology
• Reactive objects should be able to observe the
  changes in the environment and react
  automatically.
• Fixed/tuned parameters cannot support good
  reactive schemes in dynamic environments such as
  multiprocessor systems.
• Online algorithms and competitive analysis seem
  to be a promising direction.

20
Thank you!