CS848 Paper Review Form - Fall 2006

Paper Title: A method for transparent admission control and request 
scheduling in E-commerce web sites

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
 [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)
The paper proposes a Gatekeeper proxy software that sits between application
server and database server in common multi-tier e-commerce web sites. The main
contribution is that, the proxy can be transparently applied to exising web sites
without extra changes, achieving stable behavior during overload and improved 
response times by using admission control and request scheduling. The norvelty 
compared to previous research lies in its zero-modification transparency to integrate
with existing systems and the more concerns taken when calculating service cost and 
differenciating the variation in execution times of different requests. The result
shows that peak throughput is improved by up to 10% with admission control and the
success of using shortest job first(SJF) scheduling among heterogeneous requests.
The paper is accepted because of its norvelty. However, there maybe rooms for improvement
in several aspects, such as how the way system capacity is measured, the choice of request 
scheduling method and aging algorithm, locking options, and extensibility to accomodate 
the situation when application server is the bottleneck rather than the database server. 


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

S1: Good demonstration of the novelty of the paper.
S2: The success in building a transparently integratable system into prevalent web sites. 
S3: The case study upon using servlets gives good facts about java-based enterprise systems.

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

W1: NOt thorough in generalization of the method to more common cases.

10) Detailed comments for authors.

May try to evaluate some other scheduling and aging methods.
May try to improve the way to determin capacity.