Borg

E699834
Google internal system cluster management system workload scheduler

Borg is Google’s large-scale cluster management and scheduling system used to run and coordinate its internal workloads across vast data centers.

Try in SPARQL Jump to: Statements Referenced by

Statements (49)

Predicate Object
instanceOf Google internal system ⓘ
cluster management system ⓘ
workload scheduler ⓘ
architectureType centralized scheduler ⓘ
describedAt USENIX ATC 2015 NERFINISHED ⓘ
designedFor efficient resource utilization ⓘ
high reliability ⓘ
internal Google production workloads ⓘ
developer Google ⓘ
documentedIn "Large-scale cluster management at Google with Borg" ⓘ
handles hundreds of thousands of machines ⓘ
millions of jobs per week ⓘ
inspired Google Borgmon monitoring concepts ⓘ
Kubernetes NERFINISHED ⓘ
Omega NERFINISHED ⓘ
language primarily implemented in C++ ⓘ
manages Google data center resources ⓘ
large-scale compute clusters ⓘ
operatesOn Google data centers ⓘ
predecessorOf Kubernetes control plane design concepts ⓘ
provides automatic load distribution ⓘ
high availability for services ⓘ
high cluster utilization ⓘ
job lifecycle management ⓘ
job monitoring ⓘ
schedules batch jobs ⓘ
containerized workloads ⓘ
long-running services ⓘ
scope planet-scale cluster management ⓘ
securityFeatures isolation between different users and jobs ⓘ
supports affinity and anti-affinity rules ⓘ
automatic rescheduling of failed tasks ⓘ
automatic restarts ⓘ
bin packing of tasks onto machines ⓘ
fault-tolerant job execution ⓘ
health checking ⓘ
job placement constraints ⓘ
job priorities ⓘ
maintenance coordination ⓘ
multi-tenancy ⓘ
preemption ⓘ
quota enforcement ⓘ
resource isolation ⓘ
resource overcommitment ⓘ
service discovery integration ⓘ
usedBy Google NERFINISHED ⓘ
uses cgroups for resource isolation ⓘ
containers ⓘ
namespaces for isolation ⓘ

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.

Google Borg system → internalName → Borg ⓘ
subject surface form: Google Borg