Check out our new GPU Glossary! Read now
Modal logo
IntroductionCustom container images Defining ImagesPrivate registriesFast pull from registryGPUs and other resources GPU accelerationUsing CUDA on ModalReserving CPU and memoryScaling out Scaling outDicts and queuesJob processingConcurrent inputs on a single container (beta)Dynamic batching (beta)Scheduling and cron jobsDeployment Apps, Stubs, and entrypointsManaging deploymentsInvoking deployed functionsContinuous deploymentSecrets and environment variables SecretsEnvironment variablesWeb endpoints Web endpointsStreaming endpointsWeb endpoint URLsRequest timeoutsWebhook tokens (beta)Networking Tunnels (beta)Proxies (beta)Data sharing and storage Passing local dataVolumesMounting local files and directoriesStoring model weightsDataset ingestionCloud bucket mountsNetwork file systems (superseded)Sandboxes SandboxesRunning commandsNetworking and securityFile accessSnapshotsPerformance Cold start performanceMemory Snapshot (beta)Geographic latencyReliability and robustness Failures and retriesPreemptionTimeoutsTroubleshootingSecurity and privacyIntegrations Using OIDC to authenticate with external servicesConnecting Modal to your Vercel accountConnecting Modal to your Datadog accountConnecting Modal to your OpenTelemetry providerOkta SSOSlack notifications (beta)Other topics File and project structureDeveloping and debuggingModal user account setupWorkspacesEnvironmentsJupyter notebooksAsynchronous API usageGlobal variablesRegion selectionContainer lifecycle hooksParameterized functionsS3 Gateway endpoints

Geographic Latency

Modal’s worker cluster is multi-cloud and multi-region. The vast majority of workers are located in the continental USA, but we do run workers in Europe and Asia.

Modal’s control plane is hosted in Virginia, USA (us-east-1).

Any time data needs to travel between the Modal client, our control plane servers, and our workers latency will be incurred. Cloudping.com provides good estimates on the significance of the latency between regions. For example, the roundtrip latency between AWS us-east-1 (Virginia, USA) and us-west-1 (California, USA) is around 60ms.

You can observe the location identifier of a container via an environment variable. Logging this environment variable alongside latency information can reveal when geography is impacting your application performance.

Region selection

In cases where low-latency communication is required between your container and a network dependency (e.g a database), it is useful to ensure that Modal schedules your container in only regions geographically proximate to that dependency. For example, if you have an AWS RDS database in Virginia, USA (us-east-1), ensuring your Modal containers are also scheduled in Virginia means that network latency between the container and the database will be less than 5 milliseconds.

For more information, please see Region selection.