CS848 Paper Review Form - Fall 2006

Paper Title: Triage: Performance Differentiation for Storage Systems Using Adaptive Control

Author(s):Karlsson, Karamanolis and Zhu

1) Is the paper technically correct?
 [X] Yes
 [ ] Mostly (minor flaws, but mostly solid)
 [ ] No

2) Originality
 [ ] Very good (very novel, trailblazing work)
 [X] Good 
 [ ] Marginal (very incremental)
 [ ] Poor (little or nothing that is new)

3) Technical Depth
 [ ] Very good (comparable to best conference papers)
 [X] Good (comparable to typical conference papers)
 [ ] Marginal depth
 [ ] Little or no depth

4) Impact/Significance
 [ ] Very significant
 [X] Significant
 [ ] Marginal significance.
 [ ] Little or no significance.

5) Presentation
 [ ] Very well written
 [X] Generally well written
 [ ] Readable
 [ ] Needs considerable work
 [ ] Unacceptably bad

6) Overall Rating
 [ ] Strong accept (very high quality)
 [X] Accept (high quality - would argue for acceptance)
 [ ] Weak Accept (marginal, willing to accept but wouldn't argue for it)
 [ ] Weak Reject (marginal, probably reject)
 [ ] Reject (would argue for rejection)

7) Summary of the paper's main contribution and rationale
   for your recommendation. (1-2 paragraphs)

This paper looks at designing a controller for sharing the resources of a storage
system between multiple applications.  They argue that the control needs to be
adaptive to manage the overall dynamics of a realistic storage system.  The paper
presents both a nonadaptive and adaptive controller and compares the two controllers
experimentally.

The adaptive controller is designed using control theory, specifically, a direct 
self-tuning regulator.  They fit the model parameters of their controller using
least squares regression and show that if the parameters fall within certain bounds, 
then the controller will be stable.  One thing to note is that in the actual system 
the parameters do not fall enitrely within the given ranges for stability.


8) List 1-3 strengths of the paper.  (1-2 sentences each,
identified as S1, S2, S3.)

S1- Coherent design of an adaptive controller using control theory

S2- Well written with informative experiments.  Though the number of experiments
 could have been higher.

S3- Provides good insight into why an adaptive controller would be necessary.

9) List 1-3 weaknesses of the paper (1-2 sentences each,
identified as W1, W2, W3.)

W1- Didn't discuss what happens when system specs and application requirements result
in queuing delays and cache overflows.

W2- Triage system requires modification of the clients.  This would be undesirable as it
would require a re-compilation of existing applications.

W3- Why use the bands structure and how are the size of the bands selected.

10) Detailed comments for authors.

Overall, this was a good paper.  More attention to detail was required when spell
checking the document as errors were found.  Overall, the design parameters mentioned 
were well explained but some problems like queuing delay were ignored.