AC-OPF on CSV or PSS/E RAW cases, with convergence reported openly — residuals and CERC band violations surfaced, never hidden behind a pass flag
Independent solves batched onto the GPU, where parallelism actually pays. Single sparse factorisations stay on the CPU path
Build and compare scenarios against a base case without re-importing the network
Operator-driven N-k screening today via CSV line and generator outages. Automated full N-1 sweeps are the Q1 2027 target
Monitor optimization progress and solver metrics
Per-zone demand, generation, LMP, inter-zonal flows, and interstate import/export after studies
CSV case ZIP and PSS/E RAW import with air-gapped data guard validation
High-performance compute
A single AC-OPF is a sparse, sequential problem — a GPU offers little over a good CPU solver and pays real overhead moving the matrix into VRAM. The advantage appears when the work is independent and repeated: an N-1 sweep is thousands of unrelated solves, which is exactly the shape a GPU is built for. We use the GPU where that holds and say so where it does not.
How it works
AxonGrid employs a zero-trust, VPC-isolated architecture designed for high-consequence energy environments.
Core Julia compute clusters operate in air-gapped private subnets with zero internet ingress/egress. AWS communication uses private VPC Endpoints only.
AES-256 encryption with AWS KMS Customer Managed Keys (CMK). All sessions secured via TLS 1.3.
Mandatory MFA, organizational email verification, and role-based access keep control with the utility operator. Enterprise OIDC and SAML 2.0 federation supported for SSO.
Every simulation, login, and administrative action recorded in encrypted audit logs for full traceability and compliance reporting.
Trust
AxonGrid is built on a foundation of absolute data residency. To ensure the security and sovereignty of critical national infrastructure data, 100% of our computation and storage takes place within the AWS Sydney (ap-southeast-2) Region.
We chose AWS Sydney (ap-southeast-2) for maximum and continuous availability — a stable, sovereign foundation for pilot and early-production workloads close to Australian grid operators.Final product development will add local data residencies in additional regions so each deployment can meet its own jurisdictional and availability requirements.