14. Tasks in building ML model
1. Data
i. Collection
ii. Cleaning/Cleansing
2. Feature
i. Preprocessing
ii. Dictionary
3. Training
i. Algorithm
ii. Parameter tuning
iii. Evaluation
4. Others
i. Server setup
ii. Versioning
(data, parameter, model, result)
学習はツールが充実してきた
15. Tasks in serving ML service
1. High Availability
2. Management
i. Upload the latest ML model
ii. Switch a model without stopping
services
iii. Versioning models
3. Monitor
i. Load balancing
ii. Auto healing
iii. Auto scaling
iv. Performance/Results check
4. Others
i. Server setup
(development/staging/production)
ii. Integration to the existing services
iii. AB testing
iv. Managing many ML services
v. Logging
運用はツールが少ない
18. Tasks in serving ML service
1. High Availability
2. Management
i. Upload the latest ML model
ii. Switch a model without stopping
services
iii. Versioning models
3. Monitor
i. Load balancing
ii. Auto healing
iii. Auto scaling
iv. Performance/Results check
4. Others
i. Server setup
(development/staging/production)
ii. Integration to the existing services
iii. AB testing
iv. Managing many ML services
v. Logging
20. Tasks in serving ML service
1. High Availability
2. Management
i. Upload the latest ML model
ii. Switch a model without stopping
services
iii. Versioning models
3. Monitor
i. Load balancing
ii. Auto healing
iii. Auto scaling
iv. Performance/Results check
4. Others
i. Server setup
(development/staging/production)
ii. Integration to the existing services
iii. AB testing
iv. Managing many ML services
v. Logging
22. Tasks in serving ML service
1. High Availability
2. Management
i. Upload the latest ML model
ii. Switch a model without stopping
services
iii. Versioning models
3. Monitor
i. Load balancing
ii. Auto healing
iii. Auto scaling
iv. Performance/Results check
4. Others
i. Server setup
(development/staging/production)
ii. Integration to the existing services
iii. AB testing
iv. Managing many ML services
v. Logging
24. Tasks in serving ML service
1. High Availability
2. Management
i. Upload the latest ML model
ii. Switch a model without stopping
services
iii. Versioning models
3. Monitor
i. Load balancing
ii. Auto healing
iii. Auto scaling
iv. Performance/Results check
4. Others
i. Server setup
(development/staging/production)
ii. Integration to the existing services
iii. AB testing
iv. Managing many ML services
v. Logging
25. Kubernetes via Rancher
• Auto healing (Deployment/Daemonset)
• サービスが死んだら自動で起動
• Auto scaling (HorizontalAutoScaler)
• 負荷に応じて自動でPodを増減
• Rolling update
• サービスを止めずに更新
26. Tasks in serving ML service
1. High Availability
2. Management
i. Upload the latest ML model
ii. Switch a model without stopping
services
iii. Versioning models
3. Monitor
i. Load balancing
ii. Auto healing
iii. Auto scaling
iv. Performance/Results check
4. Others
i. Server setup
(development/staging/production)
ii. Integration to the existing services
iii. AB testing
iv. Managing many ML services
v. Logging
28. Tasks in serving ML service
1. High Availability
2. Management
i. Upload the latest ML model
ii. Switch a model without stopping
services
iii. Versioning models
3. Monitor
i. Load balancing
ii. Auto healing
iii. Auto scaling
iv. Performance/Results check
4. Others
i. Server setup
(development/staging/production)
ii. Integration to the existing services
iii. AB testing
iv. Managing many ML services
v. Logging
30. Tasks in serving ML service
1. High Availability
2. Management
i. Upload the latest ML model
ii. Switch a model without stopping
services
iii. Versioning models
3. Monitor
i. Load balancing
ii. Auto healing
iii. Auto scaling
iv. Performance/Results check
4. Others
i. Server setup
(development/staging/production)
ii. Integration to the existing services
iii. AB testing
iv. Managing many ML services
v. Logging
32. Tasks in serving ML service
1. High Availability
2. Management
i. Upload the latest ML model
ii. Switch a model without stopping
services
iii. Versioning models
3. Monitor
i. Load balancing
ii. Auto healing
iii. Auto scaling
iv. Performance/Results check
4. Others
i. Server setup
(development/staging/production)
ii. Integration to the existing services
iii. AB testing
iv. Managing many ML services
v. Logging