File Classification in self-* storage systems

1
File Classification in self- storage systems

• Michael Mesnier, Eno Thereska, Gregory R. Ganger,
  Daniel Ellard, Margo Seltzer

2
Introduction

• Self- infrastructure need information about
• Users
• Applications
• Policies
• Not readily provided, and cannot depend on them
  to provide them
• So? Must be learned

3
Self- storage systems

• Sub-problem of the self- structure
• Key to get hints based on what creators
  associate with their files
• File size
• File names
• Lifetimes
• Intentions determined, then decisions can be made
• Results better file organization, performance

4
Classifying Files

• Current rule-of-thumb policy selection
• Generic, not optimized
• Better distinguish classes
• Finer grained policies
• Ideally assigned at file creation
• Determine classes at creation
• Self- must learn this association
• 1) traces 2)running fs

5
So, how?

• Create model that classify based on (some
  attribs)
• Name
• Owner
• Permissions
• Must filter out irrelevant attribs
• Classifier must learn rules to do so
• Based on test set
• Then inference happens

6
The right model

• Model must be
• Scalable
• Dynamic
• Cost-sensitive (mis-prediction cost)
• Interpretable (human)
• Model selected decision trees

7
ABLE

• Attribute-based learning environment
• 1. obtain traces
• 2. make decision tree
• 3. make predictions
• Top down, until all attribs are used
• Split sample until leaves have similar file
  attribs
• After creation, query begins

8
Tests

• Based on several systems to make sure it is
  workload-independent
• DEAS03
• EECS03
• CAMPUS
• LAB
• The control MODE algorithm places all files in
  a single cluster

9
Results

• Prediction results quite good
• 90 - 100 claimed
• Clustering files by attribs are clear
• Predict that a models ruleset will converge over
  time

10
Benefits of incremental learning

• Dynamically refines model as samples become
  available
• Generally better than one-shot learners
• Sometimes one-shot performs poorly
• Ruleset of incremental learners are smaller

11
On accuracy

• More attributes chance of over-fitting
• More rules -gt smaller ratios
• Loses compression benefits
• Predictive models can have false predictions
• Can impact performance
• Things that should be in RAM is placed on disk
  instead etc.
• Solution cost functions
• Penalize errors
• Create biased tree
• System goals will need to be translated into it

12
Conclusion

• These trees provide prediction accuracies in the
  90 range
• Adaptable via incremental learning
• Continued work integration into self-
  infrastructure

13
Questions?