-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtechnical-prototype-iobserve-dash.txt
86 lines (64 loc) · 3.25 KB
/
technical-prototype-iobserve-dash.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
Technical Prototype of iObserve Dash
====================================
= Overview =
The iObserve project develops an approach integrating development and
operations (DevOps). The basic concept in iObserve is the reuse of
design-time models at run-time to support analysis and adaptation of
software systems utilizing cloud services.
This includes infrastructure, platform, and software services.
Due to the complex nature of cloud-based software systems, the
adaptation cannot be performed automatically for all contingencies.
Therefore, operator intervention is necessary.
In context of performance predictions, adaptation mechanisms rely on
past events to predict the future. This approach can be extended to
support regular seasonal behavior changes. However, it cannot predict
events which have never occured or are triggered randomly. For example,
a sales event promoting a new product or offering a substantial discount
may occur at any time. Such events may also be limited to a certain
geographic location. To mitigate this limitation, iObserve allows to
introduce behavior profiles associated with such sales event which must
be induced into the prediction and adaptation mechanism.
Both operator interventions require an interface which provides the
operator with information about the current and predicted state of the
software system, communicate adaptation goals, and allow to introduce
and modify behavior profiles. As operators are comfortable with
web-based applications, we intend to develop such application which
offers this functionality.
= Goals =
* Web-based application
* Presentation of the present architecture combined with analysis
results
* Presentation of adaptation goals (architecture candidates)
* Presentation of user behavior and workload intensities
* Editing of behavior and workload intensities
* Generation and presentation of adaptation plans
= Challenges =
* Run-time models are constantly updated. Therefore, an entity present
at one time might no longer be present when a user request details.
* The application must be extendable for future students to integrate
additional views.
* The basic navigation concept might be changed over time, but the
single visualization should easily be adapted to the new concept.
* The UI must be robust, but still use modern technologies to provide
an acceptable user experience
= Technical Requirements =
* Integration of the iObserve PCM run-time model
* Integration of TeeTime and Kieker-based analysis of iObserve
* Use of a layout engine for the graphics
* Use of Maven as build system
* Use of EMF for modeling purposes
* Java/JVM domain
= Functional Requirements =
* General dashboard with overview of the present state of the
architecture (hosts, host services etc.)
* Either adaptive layouting where single nodes can be opened or
separate view for the selected component.
* Similar view for adaptation candidates
= Project Tasks =
* Evaluation and selection of a suitable web-technology for the UI
* Integration of that technology web layouting technologies
* Overall application design
* Integration of TeeTime based analysis
* Integration of EMF technology in web context
* Refined design
* Realization of views and editors for workload and behavior models