Retail Merchandising System
Oracle Retail Merchandising System (RMS) is an Oracle Retail enterprise application that supports core merchandising functions such as item and supplier management, pricing, inventory, purchasing, and merchandising operations. If you plan to deploy RMS in Google Cloud or migrate Retail Merchandising System (RMS) from your data center to Google Cloud, you can design a secure, high-availability topology by leveraging Oracle AI Database@Google Cloud.
Learn about the reference architecture for running Oracle Retail Merchandising System in Google Cloud using Oracle AI Database@Google Cloud for the database layer and Google Compute Engineer (GCE) for the application and integration tiers. This configuration provides low latency, as Oracle AI Database services are deployed in the same Google Cloud data center.
Currently, Oracle Exadata Database Service on Dedicated Infrastructure, and Oracle Exadata Database Service on Exascale Infrastructure and Oracle Base Database Service are supported with Oracle AI Database@Google Cloud. You can check regional availability matrix to determine supported services by OCI and Google Cloud regions.
This document is intended for cloud architects, infrastructure administrators, and Oracle Retail Merchandising System administrators responsible for designing, deploying, and operating Oracle Retail Merchandising System environments. Familiarity with RMS architecture and components, Oracle AI Database, and cloud platforms including Oracle Cloud Infrastructure (OCI) and Google Cloud is recommended.
Architecture
This architecture demonstrates the deployment of Oracle Retail Merchandising System in a single Availability Zone of Google Cloud region. To support disaster recovery, deploy a similar configuration across multiple Google Cloud regions. The Databases can be configured to use Oracle Active Data Guard, and the Application stack can use rsync to synchronize the file systems across regions.
For more information on designing and implementing disaster recovery architectures, see Oracle Maximum Availability Architecture for Oracle AI Database@Google Cloud.
This architecture deploys all components within a single Google Cloud region and highlights important design considerations for Oracle Retail Merchandising System on Google Cloud with Oracle AI Database@Google Cloud.
Networking Tier
This architecture shows the deployment of Retail Merchandising Suite environment in a single availability zone of Google Cloud, ensuring low latency. The architecture consists of one Google Cloud VPC with a bastion host, load balancer, web server, Retail Merchandising System web servers, and application servers in separate subnets, and an ODB Network with Oracle AI Database. Connectivity to on-premises environments can be established using Cloud Interconnect (Dedicated/Partner) with Cloud Router for private, low-latency access (or Cloud VPN as an alternative). Compute Engine VM instances for the web server and application server can be placed in multiple instance groups.
The bastion host is deployed in a subnet with an external IP address (public), while all other Retail Merchandising System and database components reside in subnets without external IP addresses. External IP addresses can be attached to specific instances based on business or operational requirements. Secure access to private instances is provided over port 22 (SSH) through the bastion host or via Cloud Interconnect/VPN when direct connectivity to on-premises data centers is configured.
Retail Merchandising Suite application components are deployed across a single Google Cloud zone to ensure low-latency connectivity. The database is deployed in a single zone with Oracle RAC enabled by default. For regional redundancy, the database can be deployed in a second zone or region using Oracle Data Guard, providing high availability and disaster recovery at the regional level.
- When planning IP address space, account for Oracle AI Database@Google Cloud subnet requirements and address space consumption scenarios.
- Plan DNS configuration carefully, especially when using custom DNS resolvers, to support Oracle AI Database@Google Cloud DNS resolution requirements.
- For multi-region disaster recovery architectures, consider detailed network connectivity patterns and inter-region routing for Oracle AI Database@Google Cloud.
- Review backup and recovery prerequisites early in the design phase to ensure network access requirements are met.
- Use Network Security Groups (NSGs) to restrict access to database virtual machines:
- Allow SSH (port 22) access only through Google Cloud Bastion VM.
- Allow database traffic (port 1521) exclusively from approved Oracle Retail Merchandising System application subnets and authorized on-premises networks.
Bastion Host
A Google Cloud bastion host provides a secure and controlled administrative access point to the Google Cloud VPC hosting Oracle Retail Merchandising System workloads.
The bastion host is deployed in a dedicated public management subnet and acts as a jump point to reach virtual machines in private subnets that do not have external (public) IP addresses, ensuring they are not directly exposed to the public internet. By using a bastion host, the architecture maintains a single, known access point that can be centrally monitored and audited, while avoiding the need to expose public IP addresses or open inbound ports on individual virtual machines.
In this architecture, target virtual machines do not require external IP addresses. Administrative access is established using SSH (port 22) from an administrator workstation to the bastion host, and then onward from the bastion to private instances over internal IPs. Firewall rules can be scoped tightly so that inbound SSH is allowed only to the bastion host (for example, from corporate public IP ranges), and private subnets allow SSH/RDP only from the bastion subnet or bastion service account, reducing the attack surface.
As an alternative to a traditional bastion with a public IP, Google Cloud also supports Identity-Aware Proxy (IAP) TCP forwarding, which enables administrative access to private instances without opening inbound SSH/RDP from the internet. With IAP, access is authenticated and authorized using Cloud IAM, and sessions can be logged via Cloud Audit Logs. This approach further reduces exposure by keeping instances private and minimizing the need for inbound firewall rules.
The bastion host enables administrators to connect to private Oracle Retail Merchandising System components using SSH for Linux and RDP for Windows (where applicable). Connections are initiated from the administrator’s workstation and routed through the bastion (or through IAP), ensuring administrative sessions are controlled, auditable, and not directly exposed to the public network.
By centralizing administrative access and eliminating direct VM exposure, a Google Cloud bastion pattern (or IAP-based access) enhances security while maintaining operational access to private workloads.
Oracle Retail Merchandising System Application Tier
Oracle Retail Merchandising System comprises a number of solutions that form the Retail Merchandise Suite. These include the following:
- Retail Invoice Matching (ReIM):
Supports invoice validation and matching between supplier invoices, purchase orders, and receiving records. ReIM helps automate discrepancy handling (price, quantity, and terms), manage deductions and chargebacks, and streamline accounts payable processes.
- Retail Price Management (RPM):
Centralizes pricing and promotion management across the retail enterprise. RPM enables the creation, approval, and implementation of regular prices, clearances, and promotional pricing, with controls for effective dates, zones, and pricing rules to maintain consistency and compliance.
- Retail Trade Management (RTM):
Manages trade funding and supplier deals, including contracts, accruals, claims, and settlement. RTM provides visibility into vendor-funded promotions and allowances, helping retailers optimize supplier funding and ensure accurate financial reconciliation.
- Retail Sales Audit (ReSA):
Collects, validates, and audits point-of-sale (POS) transaction data to ensure accuracy and completeness before financial posting. ReSA helps detect exceptions (cash variances, transaction anomalies), supports store balancing, and improves integrity of sales and tender data used downstream.
- Retail Allocation (RA):
Optimizes allocation of inventory to stores and channels based on demand, forecasts, inventory levels, and business rules. RA supports initial allocations and replenishment-related distribution decisions to improve in-stock position and reduce over/under allocation.
- Retail Integration Bus (RIB):
Provides near real-time integration capabilities between Oracle Retail applications and external systems using message-based integration patterns. RIB supports publication and subscription of business events (for example, item, price, inventory, and purchase order updates) to keep systems synchronized.
- Bulk Data Integration (BDI):
Enables high-volume, batch-oriented data integration for foundational and operational data. BDI is commonly used for large data loads and periodic synchronizations (for example, items, suppliers, locations, costs, and hierarchies), complementing RIB’s event-driven integrations.
Additional modules commonly used with RMS are:
- Retail Store Inventory Management (SIM):
Supports store-level inventory operations such as receiving, stock counts, adjustments, transfers, returns, and store replenishment activities. SIM improves store inventory accuracy and execution, and helps align store operations with enterprise inventory and merchandising processes.
- Oracle BI Publisher (OBI PuB):
Provides pixel-perfect reporting and document generation (for example, forms, operational reports, and scheduled report bursts). In Oracle Retail environments, BI Publisher is frequently used to produce standardized reports for merchandising, finance, and store operations.
- Identity Management (IDM):
Provides centralized authentication, authorization, and user lifecycle management across retail applications. IDM supports capabilities such as single sign-on (SSO), role-based access control, and integration with corporate directories, helping enforce security policies and streamline access administration.
Database Tier
- Oracle Exadata Database Service on Dedicated Infrastructure
- Oracle Exadata Database Service on Exascale Infrastructure
- Oracle Base Database Service
The database instances are configured for high availability with Oracle Real Application Clusters (RAC) enabled. To achieve availability zone redundancy for the database, use Oracle Active Data Guard in synchronous mode to replicate the database across availability zones.
- Deploying two different ODB networks in a VPC, or
- OCI backbone connectivity using VCN peering with Local Peering Gateways.
Port 1521 must be open for communication with Oracle Active Data Guard, as Data Guard transport services use port 1521 to transmit redo log files. For detailed networking design considerations, see Maximum Availability Architecture (MAA).
Backup and Recovery
Automated database backups can be configured using Oracle Autonomous Recovery Service or OCI Object Storage, depending on the selected database service and recovery requirements.
Data Encryption
For data in transit, Oracle AI Database@Google Cloud services are accessible only through encrypted communication channels. By default, the Oracle Net client is configured to use encrypted sessions, ensuring that all database connections are protected in transit.
Oracle AI Database@Google Cloud protects data at rest using Transparent Data Encryption (TDE), which is enabled by default with no customer configuration required. TDE automatically encrypts database files, redo and undo logs, backups, and other persistent data when written to storage, and transparently decrypts the data when accessed by authorized processes. Encryption is managed using a hierarchical key model, where a master encryption key protects tablespace keys that in turn encrypt the data.
Oracle AI Database@Google Cloud supports both Oracle-managed and customer-managed key options for TDE. With Oracle-managed keys, encryption keys are generated, stored, and managed automatically by Oracle. With customer-managed keys, customers can centrally control key lifecycle management, rotation, and auditing by integrating with OCI Vault, Oracle Key Vault, or Google Cloud Key Management Service (KMS).
Cross-region Oracle Data Guard is not supported when customer-managed encryption keys are stored in Google Cloud Key Management Service (KMS).
Migration to Oracle AI Database@Google Cloud
Oracle Zero Downtime Migration (ZDM) provides multiple migration workflows for moving Retail Merchandising System databases to Oracle AI Database@Google Cloud.
- Physical Online Migration:
The physical online migration workflow supports migrations between the same database versions and platforms. This approach uses direct data transfer and the restore from service method to create the target database, avoiding the use of intermediate backup storage. Oracle Data Guard is used to keep the source and target databases synchronized, enabling a minimal-downtime migration.
- Physical Offline Migration:
The physical offline migration workflow supports migrations between the same database versions and platforms. The target database is created using Recovery Manager (RMAN) backup and restore. Google Cloud Filestore is used to provide an NFS file share for storing RMAN backup files during the migration process.
- Logical Online Migration:
The logical online migration workflow supports migrations between the same or different database versions and platforms. This workflow uses Oracle Data Pump export and import to create the target database. Google Cloud Filestore provides an NFS file share to store the Data Pump dump files. Oracle GoldenGate is used to synchronize the source and target databases, enabling a minimal-downtime migration.
- Physical Offline Migration:
The logical offline migration workflow supports migrations between the same or different database versions and platforms. The target database is created using Oracle Data Pump export and import.Google Cloud Filestore provides an NFS file share to store the Data Pump dump files used during the migration.
Components Overview
| Component | Purpose |
|---|---|
| Oracle AI Database@Google Cloud |
Oracle AI Database@Google Cloud provides Oracle Exadata Database deployed and operated in Google Cloud with native Google Cloud integration. It combines Oracle Exadata Database performance and Oracle AI Database capabilities with Google Cloud networking, security, and consumption models. The offering includes Oracle Exadata Database Service on Dedicated Infrastructure, Oracle Exadata Database Service on Exascale Infrastructure, and Oracle Base Database Service for hosting database layer for hosting the database layer.. |
| Load Balancer |
Load Balancer distributes incoming traffic across web or application servers and continuously monitors back-end health probes to send traffic only to healthy instances. This ensures even traffic distribution, high availability, and automatic failover without application. |
| Bastion |
Google Cloud Bastion Virtual Machine enables secure RDP and SSH access to virtual machines over HTTPS without requiring public IP addresses. It improves security by centralizing administrative access and reducing exposure to inbound internet threats. |
| Autonomous Recovery Service |
Autonomous Recovery Service provides automated backup, continuous data protection, and fast recovery for Oracle AI Database(s). It reduces data loss and recovery time by autonomously managing backups, validation, and restore operations. |
| Object Storage |
Object Storage provides durable, scalable storage for unstructured data using a bucket-and-object model. It is commonly used for backups, archival, and data sharing with built-in security and lifecycle controls. |
| OCI Vault |
OCI Vault provides centralized management of encryption keys and secrets using Oracle-managed HSMs. It enables strong security, key rotation, and access control for protecting data across OCI services. |
| Google Cloud Filestore |
Google Cloud Filestore provides fully managed, shared file storage for Google Cloud using the standard NFS protocol. It enables applications to access scalable, high-performance, highly available file shares without managing underlying file servers or storage infrastructure. |
| Google Cloud Key Management Service (Cloud KMS) |
Google Cloud Key Management Service (Cloud KMS) is a managed service that provides secure creation, storage, and lifecycle management for secrets, encryption keys and certificated used by enterprise applications. |
Learn More
- Support for Retail Merchandising System Applications Running in a Multicloud Environment (MOS: 3066767.1)
- Retail Merchandising System