An earlier version 1.0 can be found here: https://www.slideshare.net/xqin74/how-to-write-papers-part-1-principles/edit?src=slideview 5 Simple Steps to Write a Good Research Paper Title 1. Ask yourself these questions and make note of the answers What is my paper about? What techniques/ designs were used? Who/what is studied? What were the results? 2. Use your answers to list key words. 3. Create a sentence that includes the key words you listed. 4. Delete all unnecessary/repetitive words and link the remaining. 5. Delete non-essential information and reword the title.

1. 1

2. This is my story.

3. — Samuel Johnson, The Life of Samuel Johnson LL.D. Vol 2
3

4. 4
Novel IdeasImportant Problem Realistic Solution

5. 5

6. — Oscar Wilde, Poet and playwrights
6

7. 7
Good writing is a skill you can learn.
It’s a skill that is worth learning.

8. 8
Idea Do Research Write Paper

9. Write PaperIdea
9
Do Research

10. Idea Write Paper
10
Do Research
Forces to be clear, focused
Crystallises what we don’t understand
Opens the way to dialogue with others
1
2
3

11. Do not be
intimidated!
11

12. Idea Write Paper
12
Do Research
Writing a paper is how you develop an
idea in the first place
It usually turns out to be more interesting
and challenging that it seemed at first
Form
Idea
Turns
Out

13. The Purpose of
Your Papers

14. Why bother?
14

15. Papers
Communicate
Ideas
15
Your goal: To infect the mind of your reader
with your idea, like a virus
Papers are far more durable than programs
(think Mozart). Example, the MapReduce paper
Idea
Durable

16. Use active verbs instead
of complex noun-based
phrases, and avoid
unnecessary details.
A good research paper title
should contain key words
used in the manuscript and
should define the nature of
the study.
Known abbreviations such
as CPU can be used. Other
lesser-known abbreviations
should be left out.
Titles
Three Basic Tips on Writing
Good Titles for Research Papers
16
Reference: https://www.editage.com/insights/3-basic-tips-on-writing-a-good-
research-paper-title

17. An Aggressive and Proactive Mechanism for File-System Resource
Management Supporting I/O-intensive Applications
Titles
Keep it simple, brief and attractive:
Examples
17
Informed Prefetching and Caching
Patterson, R. Hugo, et al. Informed prefetching and caching. ACM SOSP, 1995.

18. Efficient Algorithm for Task Scheduling in Cloud Environments
Titles
Use appropriate descriptive words:
Examples
18
Adaptive Genetic Scheduling for Big-Data Applications in Cloud
Environments

19. What is my paper about? What techniques/ designs were used? Who/what is
studied? What were the results?
Titles
Five Steps to Write a Good Title
19

20. My papers studies a new prefetching and caching mechanism to
improve disk I/O performance
Titles
Answer Four Questions
20

21. To use application-disclosed access patterns (a.k.a., hints) to improve
I/O parallelisms.
Titles
Answer Four Questions
21

22. How to dynamically allocate file buffers among:
(1) prefetching hinted blocks, (2) caching hinted blocks for reuse, and
(3) caching recently unhinted blocks.
Titles
Answer Four Questions
22

23. On a single disk, where storage parallelism is unavailable and
avoiding disk access is most beneficial, informed caching reduces the
elapsed time of these same applications by up to 36% with an
average of 13% compared to informed prefetching alone.
Titles
Answer Four Questions
23

24. Hints, inform, disk accesses, storage parallelism, buffer management,
prefetching, caching, informed caching, informed prefetching
Titles
Step 2:
Use Your Answers to List Key Words
24

25. This study investigates a way of applying application-disclosed
access patterns or hints to exploit I/O parallelisms; the study shows
how to dynamically allocate file buffers among three competing I/O
demands: hinted prefetching, hinted caching, and unhinted caching.
Titles
Step 3:
Create a sentence that includes the
key words you listed.
25

26. This study investigates a way of applying application-disclosed
access patterns or hints to exploit I/O parallelisms; the study shows
how to dynamically allocate file buffers among three competing I/O
demands: hinted prefetching, hinted caching, and unhinted caching.
Titles
Step 4:
Delete all unnecessary/ repetitive
words and link the remaining.
26

27. This study investigates a way of applying application-disclosed
access patterns or hints to exploit I/O parallelisms; the study shows
how to dynamically allocate file buffers among three competing I/O
demands: hinted prefetching, hinted caching, and unhinted caching.
Informed Prefetching and Catching
Titles
Step 5:
Delete non-essential information
and reword the title.
27

28. This is a grand challenge
(see also slides 11-12)
“The main idea of this
paper is....”
“In this section we present
the main contributions of
the paper.”
Many papers contain good
ideas, but do not distil
what they are.
Your Idea
Idea：A re-usable insight,
useful to your readers
28

29. Clusters have been increasingly used for scientific and commercial
applications. In a cluster environment, scientific application distributed
their data across multiple computation nodes. In order to improve the
performance of the clusters, many issues in parallel I/O have to be
judiciously investigated. These issues include: parallel file systems, access
patterns, low-level I/O interface, scientific data libraries, and data
management.
Our experiment shows that
network is one of the potential bottlenecks in cluster-based parallel I/O.
Furthermore, the performance of the ……
Abstract
A Bad Example
29

30. In this paper, we present aggressive, proactive mechanisms that tailor file
system resource management to the needs of l/O-intensive applications.
We
have implemented informed prefetching and caching in Digital’s OSF/1
operating system and measured its performance on ……
Abstract
A Good Example
30
Patterson, R. Hugo, et al. Informed prefetching and caching. ACM SOSP, 1995.

31. Your paper should have just one “ping”:
one clear, sharp idea
Read your paper again: can you hear the
“ping”?
You may not know exactly what the ping
is when you start writing; but you must
know when you finish
If you have lots of ideas, write lots of
papers
31

32. Describe your WizWoz system.
Your reader does not have a WizWoz
Readers are primarily interested in re-
usable brain-stuff, not executable artefacts
Tao has a WizWoz system – a simulator.
32
The purpose of your
paper is not...

33. 33
Here is my idea
1 2 3 4 5
Narrative Flow of Your Paper
6

34. (1000 readers)
(4 sentences, 100 readers)
(1 page, 100 readers)
(1 page, 10 readers)
(2 pages, 10 readers)
(5 pages, 3 readers)
(1-2 pages, 10 readers)
(0.5
pages)
34
Structure of Your Paper

35. 35
Abstract
Write the abstract last
Used by program committee members to
decide which papers to read
• State the problem
• Say why it’s an interesting problem
• Say what your solution achieves
• Say what follows from your
solution

36. 36
Introduction
Describe the problem and motivations
State your contributions
Summarize the roadmap
One page

37. Use an example to
introduce the problem
Describe a
Problem

38. 38
State Your
Contributions
Write the list of contributions first
The list of contributions drives the entire
paper: the paper substantiates the claims
you have made
Reader thinks “gosh, if they can really
deliver this, that’s be exciting; I’d better
read on”

39. An
Example
Bulleted
List
Do not
leave the
reader to
guess.

40. A Better
Example
Bulleted
List
Forward
References

41. Yes and No
Contributions
should be
refutable
NO! YES!
We describe the PRE-
BUD system. It is really
cool.
…to examine how to prefetch data blocks
with maximum potential energy savings into
buffer disks, thereby reducing the number
of power-state transitions and increasing
the number of standby periods to improve
energy efficiency.
We study an energy
consumption model.
We build a new energy-saving prediction
model, based on which an energy-saving
calculation module … (see Section 4.2)
We developed a
prefetching algorithm.
We developed an energy-efficient
prefetching algorithm in the context of two
buffer disk configurations. A greedy
prefetching module … (see Section 4.1)

42. Related
Work
42
Highlight
Novelty
Summarize
Differences

43. No related
work yet!
43
Related
work
Your Reader Your Ideas

44. Related
Work First
44
Your Reader Your Ideas
Related Work
Is your idea novel?
What are your
contribution?

45. Related
Work
45
You want to convince
reviewers in the first place.
You feel this is a novel idea.

46. but FIRST convey the idea
Ideas
46
Sounds
impressive,
but …
Presenting the idea
3. The idea
Consider a bifircuated semi-lattice D, over
a hyper-modulated signature S. Suppose
pi is an element of D. Then we know for
every such pi there is an epi-modulus j,
such that pj < pi.

47. Conveying the intuition is
primary, not secondary
Once your reader has the
intuition, the reader can
follow the details (but not
vice versa)
Even if the reader skips the
details, she still takes away
something valuable
Presenting
Idea
Explain your idea as if you were
speaking to someone using a
whiteboard
47

48. Examples
48
Introduce the problem, and your idea, using
EXAMPLES
and only then present the general case
The payload of your paper

49. An Example
49

50. The body of the paper
provides evidence to
support each claim
Check each claim in the
introduction, identify the
evidence, and forward-
reference it from the claim
Evidence can be: analysis
and comparison, theorems,
measurements, case
studies
Evidence Your introduction makes claims,
then …
50

51. Related Work
51

52. The Truth
52

53. 53
Compared with the introduction section.
Do not let others compete against me
using my future work ideas.
Point out future directions that seems to
be a dead-end.
A future work I believe I can complete
sooner than the others.

54. 54
Refer to the weakness of my work
Encourage other researchers to continue
working on your project
Open issues that you plan to address

55. 55
MikTex and TexMaker
Matlab
Matlab, Visio, Draw.io, and Google docs
Software Tools

56. Identify your key idea Make your contributions
explicit
Use examples
Summary
If you remember nothing else
56