summaryrefslogtreecommitdiff
path: root/2017/roadmap-osmocon2017/roadmap.adoc
blob: ac70cd660af0f44e840b9a27e379ed5ec4418754 (plain)
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
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
Osmocom Callular Infrastructure Roadmap
=======================================
:author:	Harald Welte <hwelte@sysmocom.de>
:copyright:	2017 by Harald Welte (License: CC-BY-SA)
:backend:	slidy
:max-width:	45em

== Disclaimer

People like Dieter, Holger, Daniel and I have invested lots of our
spare time in the early years.  So funding for that came from other
paid work, and Osmocom projects were a hobby. It was a lot of fun, but
it's not a sustainable basis for the scope of the projects at this
time.

Today, our development is driven by

* what people contribute patches for, and/or
* what companies and consultants in the project get development
  contracts for (e.g. sysmocom customers)

This means in general that *very little will happen, unless somebody
commits to putting resources at it!*


== Major Roadmap Areas

* Splitting up the NITB
* Code Quality
* Testing
* Integration between 2G and 3G
* External Interfaces


== OsmoNITB

* has been at the heart of most Osmocom based cellular networks
* is generally doing a good job, but suffers from some issues
* NITB includes BSC, but BSC code also used to build OsmoBSC
** No way to test OsmoBSC without a proprietary MSC :(
* NITB includes sqlite3 based HLR
** synchronous/blocking access from single-threaded OsmoNITB
** sqlite3 (particularly via DBI) scalability issues

== RIP NITB; Long Live OsmoMSC

* New *OsmoMSC = OsmoNITB - BSC - HLR*
* New OsmoHLR (GSUP interface only)
* 3GPP A-over-IP for OsmoBSC + OsmoMSC
** New SCCP+M3UA stack, libosmo-sigtran
* M3UA based IuCS/IuPS in HNB-GW, MSC, SGSN
** used to be SUA, as it was simpler initially
* Single code base/branch for 2G and 3G

-> typical 2G setup will become *OsmoBSC + OsmoMSC + OsmoHLR*

== OsmoSTP

Signal Transfer Point, part of libosmo-sccp

* M3UA + SUA support
* Connectionless and Conenction Oriented SCCP
* Used primarily for RAN-CN interface (A, Iu) so far


== osmo_fsm + osmo_prim

osmo_fsm:: structured approach to state machines

osmo_prim:: structured approach towards primitives at SAP between layers

== osmo_fsm

Generalized finite state machine (FSM) abstraction

* pure C language, no pre-processor / code-generator
* description of states, permitted events in a state, permitted
   output states
* buit-in timer support. Default action on expiry: Destroy FSM
* parent/children relationship, allows hierarchy of FSMs

Much more maintainable and deterministic than the many implicit or
non-existing state machines in older code

== osmo_fsm so far

* Ericsson OM2000 managed objects
* Connection-Oriented SCCP
* M3UA/SUA ASP + AS state machine
* All VLR FSMs (LU, Auth, ...) in OsmoMSC
* AMR DTX in OsmoBTS

== osmo_fsm candidates

* LAPD + LAPDm code
* A-bis OML managed objects
* GSM Call Control (maybe even CAMEL-like FSMs?)

More osmo_fsm Would greatly improve code quality, testability and
maintainability

* but who wants to invest in that? Any volunteers?

== Testing

* unit test coverage of most code is poor, needs attention
** coverage of new code much better than old code
* osmo_fsm state introspection via CTRL enables better testing
* end-to-end system testing software: *osmo-gsm-tester*

== osmo-gsm-tester

* python language for managing BTSs + Modems
** supports different BTS models
** can execute suites of test cases on each software version for all BTS models
* RF cabling between BTSs and Modems to avoid RF interference

-> GOAL: Daily functional testing on all supported platforms/configs

== osmo-gsm-tester

* core python infrastructure working
* jenkins integration for testing new builds working
* modem hardware being replaced due to poor ofono support for SL8082

TODO:

* modem integration using better supported hardware
* more actual test cases beyond SMS + LU
* 3G support

== External Interfaces

* *VTY* is a human interface, not intended for consumption by software
** text output syntax not guaranteed stable
** inefficient
** commands get added during development to help developers

* *CTRL* interface is the programmatic interface
** but developers don't need it during development
** all commands so far added due to user request/requirement
*** let us know what you need exposed!
** only way to fundamentally change this is to automatically export things without extra effort

=> Roadmap is to put more effort into CTRL completeness

== Grand Unified Config Theory

Big Wishlist item:

* Unified configuration store / MIB with
** automatic VTY printer/parser generation
** automatic CTRL interface exposure
** telnet/VTY process + MIB daemon outside actual applications

But who will fund this / put resources at it?

== Integration between 2G and 3G

* so far, 2G and 3G live separate to each other
* we want integration
** cross-advertisement of neighbor cells
** inter-RAT mobility
** inter-RAT hand-over
* combined PS + CS attach
* SMS over packet switched

== Operation / Maintenance

* generation of Alarms in BTS + PCU (done)
* report via CTRL TRAP as well as A-bis OML (done)
* centralized reporting / collections of Alarms (tbd)

-> we need to generate more alarms in abnormal situations

* generation of more stat, counters / KPIs
** lots of work spent in 2016 on this alrady

-> we need aggregation/interpretation/analysis of that data

== Further Wishlist

* move SMSC out of OsmoMSC
* GSUP to MAP gateway
* billing interfaces
* LTE MME/S-GW-/P-GW
** biggest issue is lack of good FOSS asn.1 tools for C
** we either have to ditch C, or invest lots of time in tools :/

== EOF

End of File
personal git repositories of Harald Welte. Your mileage may vary