Flip subsurface uncertainty into decision-ready perception
Drilling campaigns are among the many costliest and operationally complicated actions within the vitality business. Rigs, crews, gear, and logistics are scheduled months upfront, leaving little room for delay. Each hour of downtime can carry vital prices. Geoscientists and engineers depend on reservoir fashions to grasp the subsurface and information drilling selections, typically underneath intense time stress. But the present modeling infrastructure that many corporations depend on for these vital selections was designed for smaller datasets, easier workflows, and siloed operations. As subsurface knowledge grows in scale and complexity, groups want a quicker, extra related technique to rework uncertainty into actionable selections.
Now typically obtainable, agile reservoir modeling (ARM) from SLB, built-in with Azure Information Supervisor for Power (ADME), is co-engineered for cloud scale to shut that hole. The mixed answer connects ensemble-centric reservoir modeling and simulation with a ruled knowledge basis, elastic Azure high-performance computing, and scalable storage designed for the throughput and concurrency of huge ensemble research.
Scale ensemble modeling with ARM on Azure
A single reservoir mannequin represents one believable interpretation of the subsurface. Ensemble modeling evaluates a whole lot of mannequin realizations collectively, giving asset groups a broader view of potential outcomes and a stronger foundation for understanding uncertainty and threat.
The business has lengthy acknowledged the worth of ensemble modeling for risk-aware decision-making, however adoption has been constrained by infrastructure and software program designed for smaller, sequential research. Transferring from a restricted mannequin set to a whole lot of realizations multiplies the compute, storage, knowledge motion, and outcomes that should be managed. ARM addresses this barrier by distributing realization technology and simulation throughout elastic Azure high-performance computing, whereas scalable storage supplies the capability, throughput, and parallel entry required by ensembles of various sizes.
Join reservoir workflows to a ruled knowledge basis
Constructing and updating reservoir fashions attracts on many techniques: seismic, wellbore, interpretation, and simulation knowledge, plus the functions and compute that course of them. When these items are fragmented, prospects should join and orchestrate them on their very own, and large-scale simulation turns into a disconnected, one-off train.
SLB and Microsoft mixed their respective applied sciences, ARM engineered by SLB and ADME engineered by Microsoft, to take away that burden. Relatively than a set of loosely related instruments, prospects get one built-in workflow during which every element has a transparent position:
- Azure Information Supervisor for Power is the ruled knowledge basis. OSDU® Technical Normal-aligned providers and OSDU Area Information Administration Providers (DDMS) handle seismic, wellbore, interpretation, and simulation knowledge.
- Petrel™ subsurface software program is the place engineers outline interpretation and uncertainty workflows, related on to knowledge in ADME.
- Agile reservoir modeling is the ensemble workflow engine. It orchestrates the technology of reservoir realizations and simulation runs and helps discovery and evaluation at enterprise scale.
Consequently, groups construct ensembles from trusted company knowledge and publish outcomes again for discovery and reuse, which reduces knowledge motion and improves collaboration throughout asset groups.
Put ensemble modeling into apply with a related workflow
Past integration and scale, ARM distinguishes itself by orchestrating all levels of an ensemble examine throughout modeling, simulation, knowledge administration, and evaluation. The workflow tightly connects the instruments reservoir groups already use with Azure-scale infrastructure and the ruled knowledge providers offered by Azure Information Supervisor for Power.
A walkthrough of how this works at present:
Reservoir engineers outline uncertainty workflows in Petrel software program and put together simulation fashions for supported SLB simulators, together with the Intersect™ reservoir simulator and Eclipse™ reservoir simulator. ARM generates the realizations, prepares the simulation instances, and coordinates their execution throughout the Azure setting.
Azure high-performance computing executes a number of realizations in parallel. A shared, high-performance Azure Recordsdata quantity holds mannequin inputs, grids, simulation decks, non permanent recordsdata, and outcomes so the simulation fleet can learn and write concurrently. As a result of compute capability, storage capability, throughput, and IOPS might be scaled independently, the setting might be sized for the concurrency of the ensemble reasonably than the constraints of mounted infrastructure.
As simulations full, SLB reservoir engineering providers handle the ensuing ensembles and use OSDU search capabilities on ADME to help discovery, filtering, and sorting throughout giant research. Groups can find related realizations and supported, printed outcomes with out manually navigating disconnected challenge shops.
Engineers can analyze ensembles within the reservoir engineering workspace or load them again into Petrel software program for additional interpretation, historical past matching, and reservoir engineering research. This enables groups to work within the setting and software finest suited to every job whereas sustaining continuity, with out having to step outdoors their workflow to handle orchestration.
ADME supplies entitlement-based entry management built-in with Microsoft Entra ID and helps compliance controls comparable to OSDU authorized tags. Connecting ARM workflows to this ruled knowledge basis helps keep constant entry and compliance as company subsurface knowledge is used throughout modeling and simulation. As a result of ARM workflows are built-in via Petrel software program in ADME, reservoir engineers can carry out ensemble-scale research leveraging their corporate-wide knowledge shops and publish outcomes for discovery, eliminating friction and driving collaboration. Microsoft and SLB engineering collaborated carefully to construct the mixing between Petrel software program and ARM for enterprise-scale efficiency.
For reservoir engineers and asset groups, ARM turns ensemble modeling right into a repeatable workflow reasonably than a bespoke infrastructure challenge. Groups can stress-test growth ideas towards calibrated ensembles that honor noticed manufacturing historical past, not merely run a bigger variety of fashions. As new wells are drilled and new knowledge turns into obtainable, they’ll replace their reservoir understanding with out ready for a serious mannequin rebuild. For vitality IT and knowledge leaders, the workflow connects simulation to an OSDU standard-aligned structure with constant id, governance, search, and interoperability, decreasing remoted knowledge copies and disconnected functions.
The way forward for the reservoir lifecycle
Constructing on this partnership, Microsoft and SLB are working inside the OSDU neighborhood to advance a extra cloud-native customary for reservoir knowledge. This work will assist fashions, simulation outcomes, and engineering workflows function extra persistently throughout functions, enhancing interoperability whereas decreasing dependence on proprietary codecs and silos. A shared, ruled knowledge basis additionally paves the way in which for AI and agentic experiences that may uncover knowledge, coordinate workflows, and floor insights throughout the reservoir lifecycle. Mixed with Azure-scale compute and storage, this creates a path towards extra automated and steady reservoir modeling.
Get began with ARM on Azure Information Supervisor for Power
Agile reservoir modeling on Azure Information Supervisor for Power is mostly obtainable at present. Be taught extra about agile reservoir modeling from SLB and Azure Information Supervisor for Power. Contact your Microsoft and SLB account groups to establish an preliminary reservoir workflow and talk about the trail to deployment.