3 changed files with 48 additions and 65 deletions
@ -1,85 +1,51 @@ |
|||||||
# PROJECT_STATE.md |
# PROJECT_STATE.md |
||||||
|
|
||||||
Project: OTB Cloud |
Project: OTB Cloud |
||||||
Version: v1.1.0-alpha4 |
Version: v1.1.0-alpha4.1 |
||||||
Updated: 2026-04-20 |
Updated: 2026-04-20 |
||||||
Location: /opt/otb_cloud |
Location: /opt/otb_cloud |
||||||
|
|
||||||
## Current State |
## Current State |
||||||
OTB Cloud now has a working multi-profile, multi-GPU video processing pipeline integrated into the tenant storage platform. |
OTB Cloud is now operating as a live multi-profile, dual-GPU media processing/storage platform. |
||||||
|
|
||||||
### Confirmed Working |
### Confirmed Working |
||||||
- Portal-branded OTB Cloud dashboard |
- Dashboard, device browser, and workshop flow |
||||||
- Device creation and browsing |
- File-ID based workshop queueing |
||||||
- File-ID based workshop staging |
- Multi-profile processing: |
||||||
- Device-specific Video Workspace access from dashboard |
|
||||||
- Video Workshop page |
|
||||||
- Multi-profile processing selection: |
|
||||||
- default |
- default |
||||||
- compress |
- compress |
||||||
- hq |
- hq |
||||||
- Manual rotation override option: |
- Manual rotation override |
||||||
- auto/default behavior when unchecked |
- Processed outputs written to: |
||||||
- selectable 90 / 180 / 270 override when enabled |
|
||||||
- Job queue API and job listing API |
|
||||||
- File-ID based source resolution |
|
||||||
- Output routing to: |
|
||||||
- devices/<device>/video/ |
- devices/<device>/video/ |
||||||
- Profile-specific output filenames |
|
||||||
- Completed job actions: |
- Completed job actions: |
||||||
- View |
- View |
||||||
- Send to LTS |
- Send to LTS |
||||||
- Download Output |
- Download Output |
||||||
- Delete |
- Delete |
||||||
- Failed job delete action |
- Failed job delete |
||||||
- LTS routing by file type: |
- LTS routing for: |
||||||
- lts/video |
- lts/video |
||||||
- lts/archived |
- lts/archived |
||||||
- lts/pictures |
- lts/pictures |
||||||
- Health page |
- Health page with cumulative GPU time that remains after visible job deletion |
||||||
- Lifetime processing metrics retained after visible job deletion |
- Queue summary visible in workshop: |
||||||
- Intel + AMD GPU processing both in service |
- queued jobs |
||||||
- GPU time accounting active in Health page |
- active users |
||||||
- Global video jobs route exists in codebase |
- Global video jobs functionality present |
||||||
- Processed video section exists in device browser flow |
- Intel + AMD worker threads active in production use |
||||||
|
|
||||||
### Processing / GPU Behavior |
### Current Observed GPU Behavior |
||||||
Current live behavior: |
- Both intel and amd workers are active |
||||||
- both GPUs take jobs |
- Both GPUs are successfully processing jobs |
||||||
- AMD prioritizes heavier / HQ work first |
- AMD and Intel are both taking work from the queue in live operation |
||||||
- Intel handles lighter work |
- Current observed job records still show batch_id as NULL on recent jobs, so full batch-aware fairness logic should be treated as not yet proven active |
||||||
- workers continue taking suitable jobs from the queue batch as available |
|
||||||
|
### Notes on Fair Use / Scheduler |
||||||
### Latest Proven Health State |
- Fair-use scheduler direction is defined |
||||||
Health page currently shows stable cumulative values including: |
- Per-tenant one-GPU fairness under contention is still an active implementation/testing area |
||||||
- uploaded file counts and space |
- Current system is functional and stable enough to continue development while that is finalized |
||||||
- LTS counts and space |
|
||||||
- archive counts and space |
### Recommended Next Step |
||||||
- total jobs |
Proceed to Android app update: |
||||||
- completed jobs |
- allow app-side selection for images, videos, or both |
||||||
- failed jobs |
|
||||||
- cumulative GPU time that does not zero out when workshop cards are deleted |
|
||||||
|
|
||||||
### Current Storage Layout |
|
||||||
- originals remain in device originals tree |
|
||||||
- processed outputs go to: |
|
||||||
- devices/<device>/video/ |
|
||||||
- LTS destinations include: |
|
||||||
- lts/video/ |
|
||||||
- lts/archived/ |
|
||||||
- lts/pictures/ |
|
||||||
|
|
||||||
## Known Remaining Improvements |
|
||||||
- README is now being realigned to actual live state |
|
||||||
- Global video jobs page should be fully wired into UI navigation and polished |
|
||||||
- Dashboard template still contains some mixed button class styles that should be normalized |
|
||||||
- Health page can be expanded with per-processor breakdown later |
|
||||||
- Processing metrics can be refined further into Intel/AMD/CPU buckets if desired |
|
||||||
- Output browsing UX can still be improved further with richer previewing and filtering |
|
||||||
|
|
||||||
## Recommended Next Step |
|
||||||
Proceed after alpha4 with: |
|
||||||
1. global video jobs page polish and filters |
|
||||||
2. per-processor GPU metrics split (Intel / AMD / CPU) |
|
||||||
3. scheduler documentation and/or scheduler UI visibility |
|
||||||
4. processed output browsing improvements in device view |
|
||||||
|
|||||||
Loading…
Reference in new issue