Assignment 1A: Review of two research papers
Gabriel Campeanu
Software Engineering Master
PAPER 1
Iain Bate and Ralf Reutemann, Worst-Case Execution Time Analysis for Dynamic Branch Predictos
The paper is well organized andall the important sections are present.
Title is clear, informative, not to long and presents the subject of the paper.
Abstract has a decent length, but it would be nicer if there were two more lines added. The abstract provides general outlines about the purpose of the study, the methods used, what results were found and what was concluded.
Introductionstarts with a sentence that can be improved because the formulation is not one of the best.Introduction has a good structure that includes background of the subject, current state of knowledge related to the subject, contributions, related work.This section has a good ending with a summary of how the paper is organized.
In this section, there is a sentence too long that has more than four lines length.
Main Section describeswell all the sub-sections; there are just few comments for improving them. Related Work section presents clear the relevant work with references and specifies their results but it does not presents what are the results of the paper. Method section is well explained. Equations are well numbered and referenced.
.
Summarysection is well written and presents the results of the experiments in text and in graphs.The results of the graphs are well presented in accompanied text. Graphs are well done, clear, the axes are labeled and legends are presents.
Conclusions section has a nice structure, presenting the summary of the paper, the advantages, disadvantages and limitation of the results and in the last part of the section, future work is present.
References section contains all references in a good format; all references from text are referenced.
Language
Paragraphs have, in general, a good mixture of short and long sentences, but there are few long sentences over four lines length that can produce some ambiguities. Few commas are missing. The titles of the tables from the method section should be above the table picture. In some parts of the text are used subscripts, this should be improved.
General comments
The paper is good, well presented, has a correct length, a good structure, format,all sections are put in a logical way, sentences are complete, just few minor adjustments are needed.
PAPER 2
Norman P. Jouppi, Improving Direct-Mapped Cache Performance by the Addition of a Small Fully-Associative Cache and Prefetch Buffers
Organization of the paper should be improved, the result section is embedded in method section and the length of the paper is too long. There is no Related Work section.
Title is too long and can create confusion for the reader.
Abstract section is too long, more than 250 words. The author uses an acronym in this section that should be avoided. As a rule of thumb, the abstract should be written with italic font but not mandatory. This section is too technical and rather confuses the reader than selling the paper. The abstract is focused on the methods used in the research instead of a brief description of them.
Introduction section contains a table that creates confusion for the reader because is not very clear (contains a question mark instead of a machine name and this note is not explained in a legend or in text). This section misses to presents the methods uses in this research paper, methods that are described in Abstract section. Also, the background does not presents the current situation of the problem (with references to other experiments/researches) and conclusions found at the end of this research paper are missing from this section.
In Main section there is no distinguish Related Work section and the results are in sub-sections of methods presentation.There are presented several graphs that are resulted from experiments, but author does not specifies if the experiments are from some other research papers or are done by him.The Method section is too long but well described.
Summary section does not exists separately, or better to say, is included in Method section.
Conclusions section could be shorter; the focus is put on presenting the pros of the methods discussed in Method section, and just few comments on the cons of these methods. There is a graph that shows an example of the methods stated previously, but it should be moved in Main section area, because the author already presented his conclusion with words and this is not the Result section where should this graph be put.A good point is that Future Work is discussed in the end of the section.
References section is not well written, the correct format is not respected for some references.
Language
The title of tables and figures does not have thecorrect format (the numbered table/figure name should be followed by the “.” sign). The name of the tables should be above the table figure. In the introduction section, the author uses lots of exclamation marks that should be avoided. Few commas are missing and after using the sign“%”, the author uses also the “percent” word (3.1. Miss caching).Some graphs are hard to read (symbols are over-lapping).There are many acronyms that are not explained. The technical language is good.It is not necessary to use brackets for i.e. and e.g. examples used in text.
General comments
The paper should be significantly improved, a part from abstract should be moved to Introduction, a Related Work section should be created; references and table/figures titles should follow the proper format. Also, title of few figures should be improved to be clearer.
There should be a mail address of author under the title, not a street address.