[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
[hyades-dev] Meeting minutes for January 13th Hyades Execution Team Meeting
|
Sorry for the delay, I have attached the meeting notes for the previous execution team meeting. I will provide dial-up and sametime information for this week's meeting in the morning.
Kent
Execution Team Group Weekly Meeting, 11AM Eastern, Tuesday, January 13, 2004
Attending
---------
Kent Siefkes, IBM Rational Raleigh
Joe Toomey, IBM Rational Lexington
Todd Merriweather, IBM Rational Raleigh
Ernest, IBM Rational Raleigh
David Hodges, IBM Rational Raleigh
Marius, IBM Toronto
Si Dinh, IBM Rational Lexington
Proposed Agenda
---------------
1. Quick status of execution deliveries included in Hyades 1.2 (December 19th)
2. Discuss execution-related features/requirements for the upcoming first-half 2004 Hyades releases (1.2.x, 1.3?, and
2.0(=3.0)?)
3. Discuss staffing participation for the upcoming 2004 releases
4. Discuss technical issues around having the test harness support stats model collection
Recorded Discussion/Decisons (new discussion/status preceded by *)
------------------------------------------------------------------
1. Quick status of execution deliveries included in Hyades 1.2 (December 19th)
* Everything in except datapool edit/runtime API (as expected)
Expect 1.2.1 stream in place 1/14. Datapools are planned to go in the 1.3 (head).
2. Discuss execution-related features/requirements for the upcoming first-half 2004 Hyades releases (1.2.x, 1.3?, and 3.0?)
1.3
Datapool edit & runtime API.
1.2 JUnit runner doesn't work on mainframe (ASCII/EBCDIC issue)
3.0 (previously named 2.0)
Using Serge's document (for now) as input for 3.0:
5.1 64-bit support (P1)
5.2 Test Component execution (P1)
5.3 Support for test execution from Test Harness (P1)
5.4 Test harness enhancements to support additional monitoring agents (P2)
5.5 Specialized API for direct invocation of test behaviors (P1)
5.6 Test assets deployment of test assets (P1)
*Recommendation of team to swap priorites of 5.4 and 5.5, so that 5.4 is P1 and 5.5 is P2.
* 5.6 - Priority of P1 is exposing the extension point to allow automatic deployment, and P2 for implementing automatic
deployment
3. Discuss staffing participation for the upcoming 2004 releases
Joe - plans on participating in meetings and design discussions, small bandwidth for coding
Si - plans on participating in meetings, design, and implementation, % TBD
Andrew - datapools edit/runtime API for 1.3
HTTP(Kent/Todd) - no features (other than datapools) planned for 1.3 or 2.0, so no staffing planned for new HTTP
features. Some impact of the proposed switch to fully use the trace model when it is expanded to support
protocol event data is anticipated.
* Side discussion on ICU4J and Hyades/Eclipse globalization
4. Discuss technical issues around having the test harness support stats model collection
Scapatech stats viewer/editor was checked into 1.2 (perfmon data collector there?, but no way to launch yet?)
Marius provided loader for Richard Duggan's published stat event schema.
Packages that don't use trace agent hierarchy (root element is Agent):
- Symptom database package
- Common package
- Test package
We discussed usage of the trace agent hierarchy for load testing applications, and decided it was probably OK, and
so we decided to proceed on that premise.
New Action Items
----------------
* Kent - look up previously scheduled work (1.2 or earlier) for additional candidates for 3.0
