Skip to content

Fix: check whether the requested gpu count is supported by gpu list members #1175

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
dleviminzi wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fix: check whether the requested gpu count is supported by gpu list members #1175

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
28 changes: 28 additions & 0 deletions pkg/gateway/services/stub.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,9 +7,12 @@ import (
"fmt"
"math"
"slices"
"strconv"
"strings"
"time"

"github.com/rs/zerolog/log"

"github.com/beam-cloud/beta9/pkg/abstractions/endpoint"
"github.com/beam-cloud/beta9/pkg/abstractions/function"
"github.com/beam-cloud/beta9/pkg/abstractions/taskqueue"
Expand Down Expand Up @@ -82,6 +85,15 @@ func (gws *GatewayService) GetOrCreateStub(ctx context.Context, in *pb.GetOrCrea
}, nil
}

for _, gpu := range gpus {
if !gws.multiGPUAvailable(gpu, in.GpuCount) {
return &pb.GetOrCreateStubResponse{
Ok: false,
ErrMsg: fmt.Sprintf("Multi-GPU is not currently available for %s.", gpu.String()),
}, nil
}
}

stubConfig := types.StubConfigV1{
Runtime: types.Runtime{
Cpu: in.Cpu,
Expand Down Expand Up @@ -442,3 +454,19 @@ func (gws *GatewayService) getLowCapacityGpus(gpus []types.GpuType) ([]string, e
}
return lowGpus, nil
}

func (gws *GatewayService) multiGPUAvailable(gpu types.GpuType, reqGpuCount uint32) bool {
for _, poolConfig := range gws.appConfig.Worker.Pools {
if poolConfig.GPUType == gpu.String() {
count, err := strconv.Atoi(poolConfig.PoolSizing.DefaultWorkerGpuCount)
if err != nil {
log.Warn().Msgf("Failed to parse default worker GPU count for %s: %v", gpu.String(), err)
continue
}
if count >= int(reqGpuCount) {
return true
}
}
}
return false
}
Loading