feat: record dra enabled resources in snapshot by SiorMeir · Pull Request #830 · NVIDIA/KAI-Scheduler

SiorMeir · 2026-01-04T13:01:08Z

Description

Part of DRA support in KAI-Scheduler, we add recording of DRA enabled GPU resources in snapshot taking.

ResourceSlices as well as ResourceClaims will now be recorded & analyzed with snapshot-tool

Related Issues

Fixes #

Checklist

Note: Ensure your PR title follows the Conventional Commits format (e.g., feat(scheduler): add new feature)

Self-reviewed
Added/updated tests (if needed)
Updated documentation (if needed)

Breaking Changes

Additional Notes

github-actions · 2026-01-18T16:07:27Z

Merging this branch changes the coverage (1 decrease, 2 increase)

Impacted Packages	Coverage Δ	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool	0.00% (ø)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources	60.69% (+5.45%)	👍
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info	63.04% (-3.70%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info	84.09% (+0.09%)	👍
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister	0.00% (ø)
github.com/NVIDIA/KAI-scheduler/test/e2e/suites/api/snapshot	0.00% (ø)

Coverage by file

Changed files (no unit tests)

Changed File	Coverage Δ	Total	Covered	Missed	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool/main.go	0.00% (ø)	136 (+11)	0	136 (+11)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra_node_resources.go	86.67% (+86.67%)	30 (+30)	26 (+26)	4 (+4)	🌟
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info.go	58.07% (-4.94%)	322 (+30)	187 (+3)	135 (+27)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info.go	86.43% (-0.06%)	258 (+21)	223 (+18)	35 (+3)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/data_lister_mock.go	0.00% (ø)	116 (+7)	0	116 (+7)
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/interface.go	0.00% (ø)	0	0	0
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/kubernetes_lister.go	0.00% (ø)	19 (+1)	0	19 (+1)

Please note that the "Total", "Covered", and "Missed" counts above refer to code statements instead of lines of code. The value in brackets refers to the test coverage of that file in the old version of the code.

Changed unit test files

github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra_node_resources_test.go
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info_test.go
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info_test.go
github.com/NVIDIA/KAI-scheduler/test/e2e/suites/api/snapshot/snapshot_dra_test.go
github.com/NVIDIA/KAI-scheduler/test/e2e/suites/api/snapshot/snapshot_suite_test.go

github-actions · 2026-01-20T11:41:53Z

Merging this branch changes the coverage (1 decrease, 2 increase)

Impacted Packages	Coverage Δ	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool	0.00% (ø)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources	60.69% (+5.45%)	👍
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info	63.04% (-3.70%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info	84.09% (+0.09%)	👍
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister	0.00% (ø)

Coverage by file

Changed files (no unit tests)

Changed File	Coverage Δ	Total	Covered	Missed	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool/main.go	0.00% (ø)	136 (+11)	0	136 (+11)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra_node_resources.go	86.67% (+86.67%)	30 (+30)	26 (+26)	4 (+4)	🌟
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info.go	58.07% (-4.94%)	322 (+30)	187 (+3)	135 (+27)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info.go	86.43% (-0.06%)	258 (+21)	223 (+18)	35 (+3)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/data_lister_mock.go	0.00% (ø)	116 (+7)	0	116 (+7)
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/interface.go	0.00% (ø)	0	0	0
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/kubernetes_lister.go	0.00% (ø)	19 (+1)	0	19 (+1)

Please note that the "Total", "Covered", and "Missed" counts above refer to code statements instead of lines of code. The value in brackets refers to the test coverage of that file in the old version of the code.

Changed unit test files

github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra_node_resources_test.go
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info_test.go
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info_test.go

github-actions · 2026-01-20T12:44:15Z

Merging this branch changes the coverage (1 decrease, 2 increase)

Impacted Packages	Coverage Δ	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool	0.00% (ø)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources	60.69% (+5.45%)	👍
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info	63.04% (-3.70%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info	84.09% (+0.09%)	👍
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister	0.00% (ø)

Coverage by file

Changed files (no unit tests)

Changed File	Coverage Δ	Total	Covered	Missed	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool/main.go	0.00% (ø)	136 (+11)	0	136 (+11)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra_node_resources.go	86.67% (+86.67%)	30 (+30)	26 (+26)	4 (+4)	🌟
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info.go	58.07% (-4.94%)	322 (+30)	187 (+3)	135 (+27)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info.go	86.43% (-0.06%)	258 (+21)	223 (+18)	35 (+3)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/data_lister_mock.go	0.00% (ø)	116 (+7)	0	116 (+7)
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/interface.go	0.00% (ø)	0	0	0
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/kubernetes_lister.go	0.00% (ø)	19 (+1)	0	19 (+1)

Please note that the "Total", "Covered", and "Missed" counts above refer to code statements instead of lines of code. The value in brackets refers to the test coverage of that file in the old version of the code.

Changed unit test files

github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra_node_resources_test.go
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info_test.go
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info_test.go

github-actions · 2026-01-21T05:37:33Z

📊 Performance Benchmark Results

Comparing PR (siormeir/dra-in-snapshot) vs main branch:

goos: linux
goarch: amd64
pkg: github.com/NVIDIA/KAI-scheduler/pkg/scheduler/actions
cpu: AMD EPYC 7763 64-Core Processor                
                                    │ main-bench.txt │            pr-bench.txt            │
                                    │     sec/op     │    sec/op     vs base              │
AllocateAction_SmallCluster-4           108.2m ±  0%   108.2m ±  9%       ~ (p=0.937 n=6)
AllocateAction_MediumCluster-4          135.8m ±  2%   134.7m ±  1%       ~ (p=0.394 n=6)
AllocateAction_LargeCluster-4           222.1m ± 13%   220.0m ± 22%       ~ (p=0.589 n=6)
ReclaimAction_SmallCluster-4            102.9m ±  0%   103.0m ±  0%       ~ (p=0.240 n=6)
ReclaimAction_MediumCluster-4           105.7m ±  0%   105.6m ±  0%       ~ (p=0.818 n=6)
PreemptAction_SmallCluster-4            103.8m ±  0%   103.8m ±  0%       ~ (p=0.699 n=6)
PreemptAction_MediumCluster-4           113.2m ±  0%   113.6m ±  0%  +0.37% (p=0.004 n=6)
ConsolidationAction_SmallCluster-4      113.6m ±  0%   113.8m ±  1%       ~ (p=0.240 n=6)
ConsolidationAction_MediumCluster-4     200.9m ±  1%   203.6m ±  1%  +1.32% (p=0.009 n=6)
FullSchedulingCycle_SmallCluster-4      105.2m ±  0%   105.2m ±  0%       ~ (p=0.818 n=6)
FullSchedulingCycle_MediumCluster-4     119.0m ±  1%   119.1m ±  0%       ~ (p=0.310 n=6)
FullSchedulingCycle_LargeCluster-4      157.4m ±  1%   157.9m ±  1%       ~ (p=0.240 n=6)
ManyQueues_MediumCluster-4              139.3m ±  1%   139.0m ±  1%       ~ (p=0.485 n=6)
GangScheduling_MediumCluster-4          156.5m ±  1%   157.7m ±  0%  +0.76% (p=0.004 n=6)
geomean                                 130.4m         130.4m        +0.07%

                                    │ main-bench.txt │            pr-bench.txt             │
                                    │      B/op      │     B/op       vs base              │
AllocateAction_SmallCluster-4           2.115Mi ± 0%    2.115Mi ± 1%       ~ (p=0.589 n=6)
AllocateAction_MediumCluster-4          11.70Mi ± 0%    11.70Mi ± 0%       ~ (p=0.818 n=6)
AllocateAction_LargeCluster-4           41.17Mi ± 0%    41.17Mi ± 0%       ~ (p=0.485 n=6)
ReclaimAction_SmallCluster-4            865.5Ki ± 1%    863.7Ki ± 1%       ~ (p=0.937 n=6)
ReclaimAction_MediumCluster-4           2.740Mi ± 0%    2.739Mi ± 0%       ~ (p=0.937 n=6)
PreemptAction_SmallCluster-4           1010.5Ki ± 1%   1008.5Ki ± 0%       ~ (p=0.621 n=6)
PreemptAction_MediumCluster-4           3.943Mi ± 0%    3.945Mi ± 0%  +0.05% (p=0.037 n=6)
ConsolidationAction_SmallCluster-4      5.552Mi ± 0%    5.550Mi ± 0%       ~ (p=0.937 n=6)
ConsolidationAction_MediumCluster-4     46.67Mi ± 0%    46.67Mi ± 0%       ~ (p=0.589 n=6)
FullSchedulingCycle_SmallCluster-4      1.343Mi ± 1%    1.346Mi ± 1%       ~ (p=0.310 n=6)
FullSchedulingCycle_MediumCluster-4     6.717Mi ± 0%    6.717Mi ± 0%       ~ (p=0.589 n=6)
FullSchedulingCycle_LargeCluster-4      22.53Mi ± 0%    22.53Mi ± 0%       ~ (p=0.699 n=6)
ManyQueues_MediumCluster-4              16.16Mi ± 0%    16.16Mi ± 0%       ~ (p=0.394 n=6)
GangScheduling_MediumCluster-4          16.88Mi ± 0%    16.88Mi ± 0%       ~ (p=0.485 n=6)
geomean                                 6.227Mi         6.226Mi       -0.01%

                                    │ main-bench.txt │           pr-bench.txt            │
                                    │   allocs/op    │  allocs/op   vs base              │
AllocateAction_SmallCluster-4            35.40k ± 0%   35.41k ± 0%       ~ (p=0.563 n=6)
AllocateAction_MediumCluster-4           322.0k ± 0%   322.0k ± 0%       ~ (p=0.329 n=6)
AllocateAction_LargeCluster-4            1.386M ± 0%   1.386M ± 0%       ~ (p=0.974 n=6)
ReclaimAction_SmallCluster-4             7.896k ± 0%   7.895k ± 0%       ~ (p=0.751 n=6)
ReclaimAction_MediumCluster-4            24.54k ± 0%   24.54k ± 0%       ~ (p=1.000 n=6)
PreemptAction_SmallCluster-4             10.79k ± 0%   10.79k ± 0%       ~ (p=1.000 n=6)
PreemptAction_MediumCluster-4            37.17k ± 0%   37.17k ± 0%       ~ (p=0.193 n=6)
ConsolidationAction_SmallCluster-4       72.36k ± 0%   72.38k ± 0%       ~ (p=0.223 n=6)
ConsolidationAction_MediumCluster-4      681.1k ± 0%   681.1k ± 0%       ~ (p=0.589 n=6)
FullSchedulingCycle_SmallCluster-4       20.71k ± 0%   20.71k ± 0%       ~ (p=0.816 n=6)
FullSchedulingCycle_MediumCluster-4      172.1k ± 0%   172.1k ± 0%       ~ (p=0.957 n=6)
FullSchedulingCycle_LargeCluster-4       720.8k ± 0%   720.8k ± 0%       ~ (p=1.000 n=6)
ManyQueues_MediumCluster-4               360.1k ± 0%   360.1k ± 0%       ~ (p=0.589 n=6)
GangScheduling_MediumCluster-4           590.6k ± 0%   590.6k ± 0%       ~ (p=0.485 n=6)
geomean                                  108.9k        108.9k       +0.00%

Legend

📉 Negative delta = Performance improvement (faster)
📈 Positive delta = Performance regression (slower)
p-value < 0.05 indicates statistically significant change

Raw benchmark data

PR branch:

goos: linux
goarch: amd64
pkg: github.com/NVIDIA/KAI-scheduler/pkg/scheduler/actions
cpu: AMD EPYC 7763 64-Core Processor                
BenchmarkAllocateAction_SmallCluster-4         	       9	 117753936 ns/op	 2245636 B/op	   35422 allocs/op
BenchmarkAllocateAction_SmallCluster-4         	      10	 107894770 ns/op	 2218072 B/op	   35406 allocs/op
BenchmarkAllocateAction_SmallCluster-4         	      10	 108484746 ns/op	 2219359 B/op	   35410 allocs/op
BenchmarkAllocateAction_SmallCluster-4         	      10	 108261670 ns/op	 2217232 B/op	   35402 allocs/op
BenchmarkAllocateAction_SmallCluster-4         	      10	 107989883 ns/op	 2217161 B/op	   35404 allocs/op
BenchmarkAllocateAction_SmallCluster-4         	      10	 108043073 ns/op	 2217056 B/op	   35403 allocs/op
BenchmarkAllocateAction_MediumCluster-4        	       8	 134032496 ns/op	12265869 B/op	  321995 allocs/op
BenchmarkAllocateAction_MediumCluster-4        	       8	 134641109 ns/op	12263701 B/op	  321991 allocs/op
BenchmarkAllocateAction_MediumCluster-4        	       8	 135169871 ns/op	12262869 B/op	  321990 allocs/op
BenchmarkAllocateAction_MediumCluster-4        	       8	 134323206 ns/op	12263318 B/op	  321991 allocs/op
BenchmarkAllocateAction_MediumCluster-4        	       8	 135627974 ns/op	12263202 B/op	  321992 allocs/op
BenchmarkAllocateAction_MediumCluster-4        	       8	 134674827 ns/op	12265958 B/op	  321991 allocs/op
BenchmarkAllocateAction_LargeCluster-4         	       5	 214528603 ns/op	43181963 B/op	 1386302 allocs/op
BenchmarkAllocateAction_LargeCluster-4         	       5	 215397661 ns/op	43173964 B/op	 1386299 allocs/op
BenchmarkAllocateAction_LargeCluster-4         	       5	 268169916 ns/op	43173878 B/op	 1386295 allocs/op
BenchmarkAllocateAction_LargeCluster-4         	       5	 218606730 ns/op	43179865 B/op	 1386294 allocs/op
BenchmarkAllocateAction_LargeCluster-4         	       5	 221354591 ns/op	43170166 B/op	 1386272 allocs/op
BenchmarkAllocateAction_LargeCluster-4         	       5	 230070213 ns/op	43172809 B/op	 1386285 allocs/op
BenchmarkReclaimAction_SmallCluster-4          	      10	 103016822 ns/op	  881258 B/op	    7866 allocs/op
BenchmarkReclaimAction_SmallCluster-4          	      10	 102915335 ns/op	  882471 B/op	    7884 allocs/op
BenchmarkReclaimAction_SmallCluster-4          	      10	 102993436 ns/op	  882296 B/op	    7894 allocs/op
BenchmarkReclaimAction_SmallCluster-4          	      10	 102965552 ns/op	  886376 B/op	    7896 allocs/op
BenchmarkReclaimAction_SmallCluster-4          	      10	 102942473 ns/op	  895176 B/op	    7897 allocs/op
BenchmarkReclaimAction_SmallCluster-4          	      10	 103110003 ns/op	  886424 B/op	    7896 allocs/op
BenchmarkReclaimAction_MediumCluster-4         	      10	 105571591 ns/op	 2873477 B/op	   24539 allocs/op
BenchmarkReclaimAction_MediumCluster-4         	      10	 105566130 ns/op	 2873436 B/op	   24540 allocs/op
BenchmarkReclaimAction_MediumCluster-4         	      10	 105668496 ns/op	 2869549 B/op	   24538 allocs/op
BenchmarkReclaimAction_MediumCluster-4         	      10	 105779006 ns/op	 2873480 B/op	   24539 allocs/op
BenchmarkReclaimAction_MediumCluster-4         	      10	 105707923 ns/op	 2869568 B/op	   24538 allocs/op
BenchmarkReclaimAction_MediumCluster-4         	      10	 105502915 ns/op	 2869628 B/op	   24539 allocs/op
BenchmarkPreemptAction_SmallCluster-4          	      10	 103753371 ns/op	 1036748 B/op	   10789 allocs/op
BenchmarkPreemptAction_SmallCluster-4          	      10	 103890563 ns/op	 1032757 B/op	   10787 allocs/op
BenchmarkPreemptAction_SmallCluster-4          	      10	 103779673 ns/op	 1032692 B/op	   10787 allocs/op
BenchmarkPreemptAction_SmallCluster-4          	      10	 103768346 ns/op	 1035517 B/op	   10786 allocs/op
BenchmarkPreemptAction_SmallCluster-4          	      10	 103722185 ns/op	 1028919 B/op	   10785 allocs/op
BenchmarkPreemptAction_SmallCluster-4          	      10	 103752572 ns/op	 1028684 B/op	   10785 allocs/op
BenchmarkPreemptAction_MediumCluster-4         	       9	 113991160 ns/op	 4138248 B/op	   37170 allocs/op
BenchmarkPreemptAction_MediumCluster-4         	       9	 113331226 ns/op	 4134088 B/op	   37169 allocs/op
BenchmarkPreemptAction_MediumCluster-4         	       9	 113604059 ns/op	 4138552 B/op	   37172 allocs/op
BenchmarkPreemptAction_MediumCluster-4         	       9	 113640802 ns/op	 4134075 B/op	   37169 allocs/op
BenchmarkPreemptAction_MediumCluster-4         	       9	 113730621 ns/op	 4138463 B/op	   37171 allocs/op
BenchmarkPreemptAction_MediumCluster-4         	       9	 113421722 ns/op	 4134248 B/op	   37170 allocs/op
BenchmarkConsolidationAction_SmallCluster-4    	       9	 113686214 ns/op	 5819960 B/op	   72367 allocs/op
BenchmarkConsolidationAction_SmallCluster-4    	       9	 113807468 ns/op	 5814496 B/op	   72338 allocs/op
BenchmarkConsolidationAction_SmallCluster-4    	       9	 113775885 ns/op	 5835752 B/op	   72427 allocs/op
BenchmarkConsolidationAction_SmallCluster-4    	       9	 113774869 ns/op	 5819225 B/op	   72383 allocs/op

Main branch:

goos: linux
goarch: amd64
pkg: github.com/NVIDIA/KAI-scheduler/pkg/scheduler/actions
cpu: AMD EPYC 7763 64-Core Processor                
BenchmarkAllocateAction_SmallCluster-4         	      10	 108769305 ns/op	 2220485 B/op	   35412 allocs/op
BenchmarkAllocateAction_SmallCluster-4         	      10	 108312997 ns/op	 2216677 B/op	   35404 allocs/op
BenchmarkAllocateAction_SmallCluster-4         	      10	 108182984 ns/op	 2216445 B/op	   35401 allocs/op
BenchmarkAllocateAction_SmallCluster-4         	      10	 108420089 ns/op	 2217490 B/op	   35404 allocs/op
BenchmarkAllocateAction_SmallCluster-4         	      10	 107717965 ns/op	 2219051 B/op	   35404 allocs/op
BenchmarkAllocateAction_SmallCluster-4         	      10	 108088084 ns/op	 2217212 B/op	   35403 allocs/op
BenchmarkAllocateAction_MediumCluster-4        	       8	 138620081 ns/op	12269858 B/op	  322002 allocs/op
BenchmarkAllocateAction_MediumCluster-4        	       8	 137224394 ns/op	12265295 B/op	  321990 allocs/op
BenchmarkAllocateAction_MediumCluster-4        	       8	 136318176 ns/op	12276999 B/op	  321989 allocs/op
BenchmarkAllocateAction_MediumCluster-4        	       8	 135366180 ns/op	12263197 B/op	  321991 allocs/op
BenchmarkAllocateAction_MediumCluster-4        	       8	 134055970 ns/op	12262450 B/op	  321986 allocs/op
BenchmarkAllocateAction_MediumCluster-4        	       8	 132983816 ns/op	12263568 B/op	  321991 allocs/op
BenchmarkAllocateAction_LargeCluster-4         	       6	 219046848 ns/op	43186284 B/op	 1386304 allocs/op
BenchmarkAllocateAction_LargeCluster-4         	       5	 220058699 ns/op	43171164 B/op	 1386277 allocs/op
BenchmarkAllocateAction_LargeCluster-4         	       5	 224200571 ns/op	43171836 B/op	 1386286 allocs/op
BenchmarkAllocateAction_LargeCluster-4         	       5	 225466274 ns/op	43173024 B/op	 1386289 allocs/op
BenchmarkAllocateAction_LargeCluster-4         	       4	 251886764 ns/op	43171738 B/op	 1386296 allocs/op
BenchmarkAllocateAction_LargeCluster-4         	       5	 218773033 ns/op	43173587 B/op	 1386295 allocs/op
BenchmarkReclaimAction_SmallCluster-4          	      10	 103073332 ns/op	  881220 B/op	    7866 allocs/op
BenchmarkReclaimAction_SmallCluster-4          	      10	 102912164 ns/op	  886730 B/op	    7887 allocs/op
BenchmarkReclaimAction_SmallCluster-4          	      10	 103068433 ns/op	  890170 B/op	    7898 allocs/op
BenchmarkReclaimAction_SmallCluster-4          	      10	 102886174 ns/op	  886178 B/op	    7896 allocs/op
BenchmarkReclaimAction_SmallCluster-4          	      10	 102897243 ns/op	  886300 B/op	    7896 allocs/op
BenchmarkReclaimAction_SmallCluster-4          	      10	 102906921 ns/op	  886279 B/op	    7896 allocs/op
BenchmarkReclaimAction_MediumCluster-4         	      10	 105741594 ns/op	 2878367 B/op	   24539 allocs/op
BenchmarkReclaimAction_MediumCluster-4         	      10	 105677697 ns/op	 2873461 B/op	   24539 allocs/op
BenchmarkReclaimAction_MediumCluster-4         	      10	 105601706 ns/op	 2869424 B/op	   24538 allocs/op
BenchmarkReclaimAction_MediumCluster-4         	      10	 105696241 ns/op	 2873408 B/op	   24539 allocs/op
BenchmarkReclaimAction_MediumCluster-4         	      10	 105565582 ns/op	 2873551 B/op	   24540 allocs/op
BenchmarkReclaimAction_MediumCluster-4         	      10	 105622864 ns/op	 2869528 B/op	   24538 allocs/op
BenchmarkPreemptAction_SmallCluster-4          	      10	 103733775 ns/op	 1028440 B/op	   10784 allocs/op
BenchmarkPreemptAction_SmallCluster-4          	      10	 103632682 ns/op	 1028901 B/op	   10785 allocs/op
BenchmarkPreemptAction_SmallCluster-4          	      10	 103850386 ns/op	 1036748 B/op	   10789 allocs/op
BenchmarkPreemptAction_SmallCluster-4          	      10	 103873069 ns/op	 1036679 B/op	   10788 allocs/op
BenchmarkPreemptAction_SmallCluster-4          	      10	 103772174 ns/op	 1032794 B/op	   10787 allocs/op
BenchmarkPreemptAction_SmallCluster-4          	      10	 103721118 ns/op	 1039184 B/op	   10786 allocs/op
BenchmarkPreemptAction_MediumCluster-4         	       9	 113379443 ns/op	 4134415 B/op	   37171 allocs/op
BenchmarkPreemptAction_MediumCluster-4         	       9	 113219838 ns/op	 4134070 B/op	   37169 allocs/op
BenchmarkPreemptAction_MediumCluster-4         	       9	 113198484 ns/op	 4134175 B/op	   37170 allocs/op
BenchmarkPreemptAction_MediumCluster-4         	       9	 113213692 ns/op	 4133888 B/op	   37168 allocs/op
BenchmarkPreemptAction_MediumCluster-4         	       9	 113098916 ns/op	 4129814 B/op	   37168 allocs/op
BenchmarkPreemptAction_MediumCluster-4         	       9	 113123889 ns/op	 4134070 B/op	   37169 allocs/op
BenchmarkConsolidationAction_SmallCluster-4    	       9	 113536607 ns/op	 5818921 B/op	   72357 allocs/op
BenchmarkConsolidationAction_SmallCluster-4    	       9	 113683116 ns/op	 5823051 B/op	   72390 allocs/op
BenchmarkConsolidationAction_SmallCluster-4    	       9	 113851224 ns/op	 5820228 B/op	   72390 allocs/op
BenchmarkConsolidationAction_SmallCluster-4    	       9	 113447305 ns/op	 5828436 B/op	   72356 allocs/op

github-actions · 2026-01-21T06:01:28Z

Merging this branch changes the coverage (1 decrease, 2 increase)

Impacted Packages	Coverage Δ	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool	0.00% (ø)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources	60.69% (+5.45%)	👍
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info	63.04% (-3.70%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info	84.09% (+0.09%)	👍
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister	0.00% (ø)

Coverage by file

Changed files (no unit tests)

Changed File	Coverage Δ	Total	Covered	Missed	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool/main.go	0.00% (ø)	136 (+11)	0	136 (+11)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra_node_resources.go	86.67% (+86.67%)	30 (+30)	26 (+26)	4 (+4)	🌟
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info.go	58.07% (-4.94%)	322 (+30)	187 (+3)	135 (+27)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info.go	86.43% (-0.06%)	258 (+21)	223 (+18)	35 (+3)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/data_lister_mock.go	0.00% (ø)	116 (+7)	0	116 (+7)
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/interface.go	0.00% (ø)	0	0	0
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/kubernetes_lister.go	0.00% (ø)	19 (+1)	0	19 (+1)

Please note that the "Total", "Covered", and "Missed" counts above refer to code statements instead of lines of code. The value in brackets refers to the test coverage of that file in the old version of the code.

Changed unit test files

github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra_node_resources_test.go
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info_test.go
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info_test.go

github-actions · 2026-01-21T11:51:59Z

Merging this branch changes the coverage (1 decrease, 2 increase)

Impacted Packages	Coverage Δ	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool	0.00% (ø)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources	60.69% (+5.45%)	👍
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info	63.04% (-3.70%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info	84.09% (+0.09%)	👍
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister	0.00% (ø)

Coverage by file

Changed files (no unit tests)

Changed File	Coverage Δ	Total	Covered	Missed	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool/main.go	0.00% (ø)	136 (+11)	0	136 (+11)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra_node_resources.go	86.67% (+86.67%)	30 (+30)	26 (+26)	4 (+4)	🌟
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info.go	58.07% (-4.94%)	322 (+30)	187 (+3)	135 (+27)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info.go	86.43% (-0.06%)	258 (+21)	223 (+18)	35 (+3)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/data_lister_mock.go	0.00% (ø)	116 (+7)	0	116 (+7)
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/interface.go	0.00% (ø)	0	0	0
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/kubernetes_lister.go	0.00% (ø)	19 (+1)	0	19 (+1)

Please note that the "Total", "Covered", and "Missed" counts above refer to code statements instead of lines of code. The value in brackets refers to the test coverage of that file in the old version of the code.

Changed unit test files

github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra_node_resources_test.go
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info_test.go
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info_test.go

pkg/scheduler/api/node_info/node_info.go

pkg/scheduler/cache/cluster_info/cluster_info.go

pkg/common/resources/dra_node_resources.go

pkg/scheduler/cache/cluster_info/cluster_info.go

pkg/scheduler/api/node_info/node_info.go

github-actions · 2026-01-25T17:30:18Z

Merging this branch changes the coverage (2 decrease, 1 increase)

Impacted Packages	Coverage Δ	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool	0.00% (ø)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources	63.89% (+8.64%)	👍
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info	63.04% (-3.70%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info	83.49% (-0.51%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister	0.00% (ø)

Coverage by file

Changed files (no unit tests)

Changed File	Coverage Δ	Total	Covered	Missed	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool/main.go	0.00% (ø)	136 (+11)	0	136 (+11)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra.go	87.69% (-0.54%)	65 (-3)	57 (-3)	8	👎
github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra_node_resources.go	97.50% (+97.50%)	40 (+40)	39 (+39)	1 (+1)	🌟
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info.go	58.07% (-4.94%)	322 (+30)	187 (+3)	135 (+27)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info.go	85.39% (-1.10%)	267 (+30)	228 (+23)	39 (+7)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/data_lister_mock.go	0.00% (ø)	123 (+14)	0	123 (+14)
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/interface.go	0.00% (ø)	0	0	0
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/kubernetes_lister.go	0.00% (ø)	30 (+12)	0	30 (+12)

Please note that the "Total", "Covered", and "Missed" counts above refer to code statements instead of lines of code. The value in brackets refers to the test coverage of that file in the old version of the code.

Changed unit test files

github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra_node_resources_test.go
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info_test.go
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info_test.go

pkg/scheduler/cache/cluster_info/cluster_info.go

pkg/scheduler/api/node_info/node_info.go

pkg/common/resources/dra_node_resources.go

pkg/scheduler/api/node_info/node_info_test.go

github-actions · 2026-01-26T15:54:18Z

Merging this branch will decrease overall coverage

Impacted Packages	Coverage Δ	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool	0.00% (ø)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources	54.29% (-0.96%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api	27.45% (ø)
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info	66.13% (-0.62%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info	83.57% (-0.43%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister	0.00% (ø)

Coverage by file

Changed files (no unit tests)

Changed File	Coverage Δ	Total	Covered	Missed	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool/main.go	0.00% (ø)	136 (+11)	0	136 (+11)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra.go	87.69% (-0.54%)	65 (-3)	57 (-3)	8	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/cluster_info.go	0.00% (ø)	21	0	21
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info.go	62.16% (-0.85%)	296 (+4)	184	112 (+4)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info.go	85.60% (-0.89%)	257 (+20)	220 (+15)	37 (+5)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/data_lister_mock.go	0.00% (ø)	123 (+14)	0	123 (+14)
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/interface.go	0.00% (ø)	0	0	0
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/kubernetes_lister.go	0.00% (ø)	30 (+12)	0	30 (+12)

Please note that the "Total", "Covered", and "Missed" counts above refer to code statements instead of lines of code. The value in brackets refers to the test coverage of that file in the old version of the code.

Changed unit test files

github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info_test.go

github-actions · 2026-01-27T09:44:27Z

Merging this branch will decrease overall coverage

Impacted Packages	Coverage Δ	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool	0.00% (ø)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources	54.29% (-0.96%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api	27.45% (ø)
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info	66.13% (-0.62%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info	83.57% (-0.43%)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister	0.00% (ø)

Coverage by file

Changed files (no unit tests)

Changed File	Coverage Δ	Total	Covered	Missed	🤖
github.com/NVIDIA/KAI-scheduler/cmd/snapshot-tool/main.go	0.00% (ø)	136 (+11)	0	136 (+11)
github.com/NVIDIA/KAI-scheduler/pkg/common/resources/dra.go	87.69% (-0.54%)	65 (-3)	57 (-3)	8	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/cluster_info.go	0.00% (ø)	21	0	21
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/api/node_info/node_info.go	62.16% (-0.85%)	296 (+4)	184	112 (+4)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info.go	85.60% (-0.89%)	257 (+20)	220 (+15)	37 (+5)	👎
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/data_lister_mock.go	0.00% (ø)	123 (+14)	0	123 (+14)
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/interface.go	0.00% (ø)	0	0	0
github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/data_lister/kubernetes_lister.go	0.00% (ø)	30 (+12)	0	30 (+12)

Please note that the "Total", "Covered", and "Missed" counts above refer to code statements instead of lines of code. The value in brackets refers to the test coverage of that file in the old version of the code.

Changed unit test files

github.com/NVIDIA/KAI-scheduler/pkg/scheduler/cache/cluster_info/cluster_info_test.go

SiorMeir force-pushed the siormeir/dra-in-snapshot branch 2 times, most recently from cd621b6 to 34e9d77 Compare January 18, 2026 15:38

SiorMeir force-pushed the siormeir/dra-in-snapshot branch from 34e9d77 to 6ed2baa Compare January 20, 2026 11:07

SiorMeir changed the title ~~feat: record dra enabled resources in snapshot~~ feat(snpashot): record dra enabled resources in snapshot Jan 20, 2026

SiorMeir force-pushed the siormeir/dra-in-snapshot branch from 6ed2baa to 08367d0 Compare January 20, 2026 12:13

SiorMeir changed the title ~~feat(snpashot): record dra enabled resources in snapshot~~ feat: record dra enabled resources in snapshot Jan 20, 2026

SiorMeir force-pushed the siormeir/dra-in-snapshot branch from 20cce47 to 08367d0 Compare January 21, 2026 05:30

SiorMeir marked this pull request as ready for review January 21, 2026 05:30

NVIDIA deleted a comment from github-actions bot Jan 21, 2026

enoodle reviewed Jan 21, 2026

View reviewed changes

enoodle reviewed Jan 25, 2026

View reviewed changes

pkg/scheduler/cache/cluster_info/cluster_info.go Outdated Show resolved Hide resolved

pkg/scheduler/api/node_info/node_info.go Outdated Show resolved Hide resolved

davidLif reviewed Jan 26, 2026

View reviewed changes

pkg/scheduler/api/node_info/node_info_test.go Outdated Show resolved Hide resolved

SiorMeir added 6 commits January 26, 2026 17:18

feat: record DRA gpus in snapshot

acaafd2

docs: CHANGELOG

b161991

refactor: remove redundent comments

2d1a362

fix: CR comments and refactor

6910838

feat: add ResourceClaim lister

4f52337

fix: CR comments, remove dead code, simplify DRA GPUs

92b0a7f

SiorMeir force-pushed the siormeir/dra-in-snapshot branch from af23e6d to 92b0a7f Compare January 26, 2026 15:18

enoodle approved these changes Jan 27, 2026

View reviewed changes

Merge branch 'main' into siormeir/dra-in-snapshot

1224077

SiorMeir added this pull request to the merge queue Jan 27, 2026

Merged via the queue into main with commit 3902efe Jan 27, 2026
8 checks passed

SiorMeir deleted the siormeir/dra-in-snapshot branch January 27, 2026 10:19

faizan-exe pushed a commit to faizan-exe/KAI-Scheduler that referenced this pull request Feb 1, 2026

feat: record dra enabled resources in snapshot (NVIDIA#830)

a0061a0

Conversation

SiorMeir commented Jan 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Related Issues

Checklist

Breaking Changes

Additional Notes

Uh oh!

github-actions bot commented Jan 18, 2026

Merging this branch changes the coverage (1 decrease, 2 increase)

Changed files (no unit tests)

Changed unit test files

Uh oh!

github-actions bot commented Jan 20, 2026

Merging this branch changes the coverage (1 decrease, 2 increase)

Changed files (no unit tests)

Changed unit test files

Uh oh!

github-actions bot commented Jan 20, 2026

Merging this branch changes the coverage (1 decrease, 2 increase)

Changed files (no unit tests)

Changed unit test files

Uh oh!

github-actions bot commented Jan 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

📊 Performance Benchmark Results

Legend

Uh oh!

github-actions bot commented Jan 21, 2026

Merging this branch changes the coverage (1 decrease, 2 increase)

Changed files (no unit tests)

Changed unit test files

Uh oh!

github-actions bot commented Jan 21, 2026

Merging this branch changes the coverage (1 decrease, 2 increase)

Changed files (no unit tests)

Changed unit test files

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

github-actions bot commented Jan 25, 2026

Merging this branch changes the coverage (2 decrease, 1 increase)

Changed files (no unit tests)

Changed unit test files

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

github-actions bot commented Jan 26, 2026

Merging this branch will decrease overall coverage

Changed files (no unit tests)

Changed unit test files

Uh oh!

github-actions bot commented Jan 27, 2026

Merging this branch will decrease overall coverage

Changed files (no unit tests)

Changed unit test files

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

SiorMeir commented Jan 4, 2026 •

edited

Loading

github-actions bot commented Jan 21, 2026 •

edited

Loading