x/hpc — HPC module
The x/hpc module extends the marketplace to batch supercomputing. Instead
of leasing long-running services, tenants submit HPC jobs — batch workloads
with resource, walltime, and partition requirements — that execute on
providers’ existing cluster schedulers: SLURM, MOAB, or Open OnDemand.
On the provider side, the daemon’s HPC integration polls on-chain jobs, dispatches them through native scheduler adapters, tracks lifecycle with configurable concurrency limits and timeouts, and recovers state crash-safely. Usage batches flow into the same settlement pipeline as cloud workloads.
What it does
Section titled “What it does”The module carries the batch model natively on chain:
- Jobs, not containers. On-chain jobs record resource, walltime, and partition requirements — the batch model rather than a container shim.
- Clusters and offerings. Operators register clusters, expose partitions and job classes as offerings, and update node metadata as the cluster changes.
- Templates. Workload templates can be created, approved, rejected, deprecated, and revoked, with approved templates searchable by publisher and type.
- Disputes and audit. Job events, security events, and usage feed a dedicated audit log in the daemon, and job disputes are flagged and resolved on chain.
Why it exists
Section titled “Why it exists”University and national-lab clusters run some of the world’s most valuable capacity at partial utilization, but their operating model — batch schedulers, allocations, walltime — doesn’t map onto container leases. A dedicated HPC module lets that capacity join the marketplace without re-platforming the cluster.
Primary objects
Section titled “Primary objects”| Concept | Definition |
|---|---|
| HPC job | A batch workload with resource, walltime, and partition requirements, executed via a cluster scheduler. |
| Scheduler adapter | The daemon component that speaks natively to SLURM, MOAB, or Open OnDemand. |
| Partition | A scheduler-level subdivision of a cluster that can be exposed to the marketplace with its own configuration. |
Core flow
Section titled “Core flow”- Submit — A job goes on-chain. Resource, walltime, and partition requirements are recorded as an HPC job through the same exchange economics.
- Dispatch — The adapter dispatches. The daemon polls on-chain jobs and dispatches through the native scheduler with configurable limits and timeouts.
- Track — Lifecycle is tracked. Polling, dispatch, tracking, and crash-safe recovery — with job, security, and usage events in a dedicated audit log.
- Settle — Usage batches settle. Metered batches flow into the standard usage-settlement pipeline from tenant escrow.
Messages
Section titled “Messages”| Message | Purpose |
|---|---|
MsgRegisterCluster | Registers a cluster with the marketplace. |
MsgUpdateCluster | Updates a registered cluster’s configuration. |
MsgDeregisterCluster | Removes a cluster from the marketplace. |
MsgCreateOffering | Creates an offering for exposed cluster capacity. |
MsgUpdateOffering | Updates an existing HPC offering. |
MsgSubmitJob | Submits an HPC job with resource, walltime, and partition requirements. |
MsgSubmitJobFromTemplate | Submits a job derived from a workload template. |
MsgReportJobStatus | Reports scheduler status for a job. |
MsgCancelJob | Cancels a submitted job. |
MsgFlagDispute | Raises a dispute over a job. |
MsgResolveDispute | Resolves a flagged job dispute. |
MsgUpdateNodeMetadata | Updates metadata for a cluster’s nodes. |
MsgCreateWorkloadTemplate | Publishes a workload template. |
MsgUpdateWorkloadTemplate | Updates a published workload template. |
MsgApproveWorkloadTemplate | Approves a submitted workload template. |
MsgRejectWorkloadTemplate | Rejects a submitted workload template. |
MsgDeprecateWorkloadTemplate | Marks a workload template as deprecated. |
MsgRevokeWorkloadTemplate | Revokes a previously approved workload template. |
MsgUpdateParams | Governance-only update of module parameters. |
Queries
Section titled “Queries”| Query | Purpose |
|---|---|
Cluster / Clusters / ClustersByProvider | Clusters individually, as a list, or filtered by provider. |
NodesByCluster | The node inventory of one cluster. |
Offering / Offerings / OfferingsByCluster | Offerings individually, as a list, or filtered by cluster. |
Job / Jobs | A single job, or the job list. |
JobsByCustomer / JobsByProvider | Jobs filtered to the customer or provider side. |
JobAccounting | Accounting records for a job’s resource usage. |
SchedulingDecision / SchedulingDecisionByJob | The recorded scheduling decision, by ID or by job. |
SchedulingMetrics | Aggregate scheduling metrics. |
NodeMetadata | Metadata recorded for a cluster’s nodes. |
WorkloadTemplate / WorkloadTemplates | A single workload template, or the template list. |
WorkloadTemplatesByPublisher / WorkloadTemplatesByType | Templates filtered by publisher or by type. |
WorkloadTemplateUsage | Where a workload template has been used. |
ApprovedWorkloadTemplates | Workload templates currently approved for use. |
SearchWorkloadTemplates | Workload templates matching a search. |
Params | Current module parameters. |
Module interactions
Section titled “Module interactions”| Module | How they interact |
|---|---|
| x/market | HPC capacity is offered and priced through the same exchange economics. |
| x/settlement | HPC job usage settles through the standard usage-settlement pipeline. |
| x/provider | Cluster operators register as providers with HPC-specific attributes. |
| x/audit | Scheduler-level audit logging complements on-chain audit attestations. |
Related documentation
Section titled “Related documentation”- HPC — where batch supercomputing sits in the protocol.
- HPC Jobs — the tenant view of submitting and tracking jobs.
- HPC Operations — running clusters for the marketplace.
- HPC Workload Templates — publishing and approving reusable job definitions.
- On-Chain Module Map — all 27 modules at a glance.
Do clusters need re-platforming?
Section titled “Do clusters need re-platforming?”No. The scheduler, partitions, and auth stay as they are — only the partitions and job classes a facility exposes become schedulable from the market.
How is HPC usage priced?
Section titled “How is HPC usage priced?”Through the same exchange economics as the rest of the marketplace. Facilities set pricing per partition and job class, and jobs are paid from tenant escrow under the governed settlement fee policy.
What audit trail does a job leave?
Section titled “What audit trail does a job leave?”Two layers: the daemon’s dedicated audit log for job, security, and usage events, plus the on-chain record — job, usage batches, settlement — validated by consensus.