Subject: Number of paper (C429) followed by characters "JGSI Review"
Resent-Date: Thu, 30 Sep 1999 23:20:01 -0400
Resent-From: Geoffrey Fox <gcf@npac.syr.edu>
Resent-To: p_gcf@npac.syr.edu
Date: Sat, 25 Sep 1999 02:03:02 +0900
From: "ISHIZAKI, Kazuaki" <ishizaki@trl.ibm.co.jp>
To: gcf@npac.syr.edu

I recommend this paper will be acceptable.

 1) INSIGHT:
The authors describe how to build a performance critical application using
Java.

 2) INTEREST:
The paper describes the performance bottlenecks in Java Core Libraries.
Especially, there are bottlenecks in excessive synchronization, excessive
heap allocation (lack of reuse).

 3) NOVELTY:
The paper shows the bottleneck in Java Core Library. But, it is not
fudamental breakthrough.

 4) SOUNDNESS:
The paper seems sound.

 5) TECHNIQUE:
To boost the performance, the author investigate a number of
synchronizations and allocations base on a profiler and improve the class
incrementally. This approach is good engineering.

 6) PRESENTATION:
The presentation is fine.

 7) COMMENTS:
In section 2 "Our Development Tools", I think that it is not a contribution
of the paper. I suggest the section is shrunk. Then, section 7 that is
interested can be extended.

If possible, in section 4.3, please show the improvement on each step.

---
Kazuaki Ishizaki
IBM Japan, Tokyo Research Laboratory
Mail : ishizaki@trl.ibm.co.jp
TEL: ++81-46-273-4664, FAX: ++81-46-274-4282