CS848 Paper Review Form - Fall 2006

Paper Title: SVL: Storage virtualization engine leveraging DBMS technology

Author(s):

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
 [ ] Generally well written
 [X] 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 proposes a storage management system, SVL, which leverages DBMS 
technology to support block storage management. It shows how block virtualization 
can be achieved by mapping logical blocks to physical blocks using tables, 
table spaces, table space maps etc. and also how existing DBMS functionalities 
can be exploited to support data encryption and compression. Finally the paper 
implements SVL specific optimizations aimed to achieve performance goals of a storage 
system and ends with an experimental evaluation of the work.

The idea presented is novel, each step is accompanied with solid technical details 
and transition from one section to the next is smooth. Another contribution of this 
work is devising a new index access method for variable length blocks. Excellent work 
overall.

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

S1: Approach is very strong, with solid technical details and empirical results.

S2: Uses an existing off-the-shelf DBMS, thus minimizing development efforts.

S3: Presents various optimizations to reduce CPU, IO and communication overhead. 


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

W1: The paper assumes a lot of technical background on operating systems, storage 
systems and DBMS on part of the reader, which makes it rather difficult to comprehend 
for an average reader.

W2: Imposes a limitation on client-server to reside on the same controller to reduce 
communication overhead by using shared-memory.

W3: The results of experimental study may be a bit flawed in the sense that it uses 
percentages to compare the performance with VSD. Saying that the performance gap is ‘only’ 
18%, is not justified because a) the actual overhead depends on the work-load under consideration 
b) this 18% overhead might not be acceptable for a real time storage system.

10) Detailed comments for authors.

- An extension to the experimental study could be to show how SVL indexing mechanisms 
behave with excessive insert/delete work-loads.

- The paper talks about blocks spanning multiple pages but does not give necessary 
details of how that would be managed.