Rally RoadMap
Comments
 Share
The version of the browser you are using is no longer supported. Please upgrade to a supported browser.Dismiss

 
Comment only
 
 
Still loading...
ABCDEFGHIJKLMNOPQRSTUVWXYZAA
1
vDescriptionResources ETALong description
2
safe-for-production
Rally must be absolutelly safe to use against proudction clouds.

allocatedn/a
It means next:
* Disaster cleanup. Even if server with running Rally was destroyed in the middle new
instance of Rally on new service will know how to cleanup everything.
* Rally deletes all own resources and doesn't delete anything else
* Rally for most of benchmarks won't need admin user at all
3
scalability Rally must have scalable engine and storage systemallocatedn/a
* Reports and Storing results should be scalable
* Benchmarks should be able to run forever (millions of iterations)
* Single task should be able to contain threads
*
4
cloud-certification
The idea is to have 1 parametrized Rally input task that checks
that everything works with required performance under expected
load at expected scale.
allocatedn/a
Input Rally task will accept very high level arguments.
E.g. amount of users/tenants/controllers/computes, specific flavors/images and purpose
of the cloud (HPCC/DevTesting/PaaS/SaaS/IaaS/Hosting/...)
5
dataplane-testing
Testing VM/Disks perfromance, and running benchmark tools
inisde the cloud
allocatedn/a
It means next:

* Improve scenario output format
* Improve reports to display this data
* Implement base plugins that simplifes porting benchmarking tools to the cloud
* Port existing benchmarking tools to Rally
6
better-reports
Various improvments in graphs to make them more
understandable and self-sufficient
allocatedn/a
* More various graphs
* Add benchmark description
* Add infomration about enviorment
7
trends-analyze-reportsAnalyzing results of the same benchmark(s) over the time requiredn/a
Such results like: Avg Duration, Amount of exceptions, SLA failures
and rest can be used to build trends. So we can fetch tasks by tags
and build such reports
8
results-comparisonReport with detailed comparions of two runs of Rallyrequiredn/a
This is very important and simplifes life for everybody that tries to
tune configuration or code
9
framework-generalization
Rally should be common tool independet from OpenStack,
that can benchmark anything including OpenStack
requiredn/a
* All hardcoded to OpenStack places should be refactored
* Restructurization of project to make it simple to split plugins from the framework
* Splitting repo into 2:
** rally with framweork
** rally-openstack-plugins
10
plugin-managment-system
Rally should have plugin managment system e.g. http://npmjs.com
requiredn/a
* Rally CLI for manage plugins (search/install/remove)
* There should be a service for uploading/sharing/rating plugins
11
flexible-benchmark-engine
A lot of real life workflow can't be generated by current
Rally benchmark engine
allocatedn/a
* Multi scenario load generation (required for real life testing)
* Resturcturization of way how rally stores data
* Better input task format:
** Strict order of execution
** Better benchmark naming & descriptions
12
ha-testing
Run in parallel descrutive scenario with normal scenario
will allow us to check that HA actually works and fully auotmate
HA testing
requiredn/a* Implement everything in flexible-benchmark-engine
* Implement HA tests
13
as a serviceRally deamon that has HTTP API
14
api for raw data processingRally raw data processing API for custom report genration
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
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Loading...
 
 
 
High Level Road Map
Long term road map
Release Schedule
Implemented Kilo
Implemented Liberty