ABCDEFGHIJKLMNOPQRSTUVWXYZAAABAC
1
ValueFeatureOwnerPriorityPartner / DependencyNotesUser story
2
2016/Q4
3
Runtime portabilityContainer Runtime Interfaceyjhong@googleP0Lock down v1alpha1 for CRI including logging, monitoring, attach, etc.
4
CRI client-side implementation in kubeletyjhong@google, feiskyer@hyperP0http://hypercontainer.io/
5
Docker integration via CRIyjhong@googleP0Monitoring, logging, and exec/attach/port-forward are handled separately by following items
6
Hyperd support via CRIfeiskyer@hyperP1http://hypercontainer.io/
7
rktleteuan.kemp@coreos, yifan-gu@coreosP1coreos
8
OCI compatible runtimemrupal@redhatP1Red Hat
9
CRI network pluginmixia@googleP0sig-cluster
10
CRI logging handlingtmrts@coreosP0coreosGoal for 1.5 release?
11
CRI exec/attach/port-forward (TBD)yifan-gu@coreos, decarr@redhat (he will find another redhatter), tsclair@ from node teamP1
12
Node release qualityGCI as default for nodevishnu@google, stclair@googleP0GCIstclair@ will help GCI effort once AppArmore work is clear
13
Node e2e test coveragelantaol@googP1
14
Speed up cluster creationPre-download Kubernetes imagesameyd@google, maisem@googleP0GCI, cluster-lifecycle
15
Better node upgradeIn-place update Kubelet? from node team, ? from gci team, robertbailey@google P1sig-cluster-lifecycle
16
Simpler node setupGA: dynamic config Kubeletmtaufen@googleP0
17
Better security isolationAppArmor support enhancementstclair@google
P0- cAdvisor profile status report
- Kubelet profile management
- System component profiles
18
Better resource provisioningNode allocatable Rollout to GKEdawnchen@googleP0GCI?
19
Alpha: Per pod cgroupderekcarr@redhat, ? from node teamP0RedhatMake sure working as an alpha feature on both systemd and non-systemd
20
Improved node reliabilityEviction uses memcg notifications instead of polling cAdvisorsjenning@redhat.comP1Red Hatsee: https://github.com/kubernetes/kubernetes/issues/31362
21
Out-of-inode eviction understands per-container inode usagesdashpole@google.comP1Google
22
Stateful workload supportGA: init containererictune@goog, dawnchen@P1Red Hat (clayton)
23
Better security isolationusernspweil@redhatP0
24
More resource typeAlpha: GPU supportHui-Zhi, jlewi@googleP2GCI, Scheduler(?)Many users asked this feature including jlewi@ is for a while. Jeremy agreed to review the proposal & implementation together with sig-node team.
25
Scheduler extensibilityOpaque integer resource accounting{Connor, Balaji}@IntelP0Intel, SchedulerBased on PoC work, may need to land a small patch to Kubelet to enable editing capacity in node status. This feature is currently marked P0 in sig-sched plan for v1.5, therefore mirroring here to represent node dependency.
26
User experienceOptional explicit QoS class override{Connor, Balaji}@IntelP2Intel, APISee issue https://github.com/kubernetes/kubernetes/issues/32468
27
Secure infrastructureKubelet API security brandon.philips@coreos (Please help to define the issue and scope here :-))P0sig-security, sig-cluster-lifecycleSee: https://github.com/kubernetes/kubernetes/issues/11816
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