The following sections describe the architecture components that are used for the Dell Validated Design for Manufacturing Edge with PTC .
Home > Edge > Manufacturing Edge > Guides > Dell Validated Design for Manufacturing Edge with PTC - Design Guide > Architecture components
The following sections describe the architecture components that are used for the Dell Validated Design for Manufacturing Edge with PTC .
VxRail is a preconfigured VMware hyperconverged infrastructure, or HCI. An HCI system includes, at a minimum, compute, software-defined storage, and virtualized networking and can run on commercial off-the-shelf servers. The underlying resources are abstracted and pooled together which allows them to be dynamically allocated to applications running in VMs or containers. The components that make up a VxRail configuration are as follows:
VxRail hardware consists of Dell PowerEdge servers. The servers are configured with Intel Xeon or AMD EPYC processors. Memory capacity ranges from 64 GB to 6,144 GB per node, depending on the model. The VxRail nodes are enclosed in a one-node, single-server system, with each node having one, two, or four multicore processors and either all-flash solid-state disks (SSDs) or a hybrid mix of flash SSDs and hard disk drives (HDDs). The nodes form a networked cluster with a minimum of two nodes or three nodes for scale-out clusters with a maximum of 64 nodes. Each server node includes the following technology:
Configuration is highly customizable and designed to meet any HCI requirements.
A typical implementation uses one or more 10 GbE Top-of-Rack (ToR) switches to connect to each node in a VxRail Cluster. For smaller environments, 1 GbE switches may also be used. While the customer typically provides the ToR switches, Dell Technologies offers Ethernet switches for management. Two ToR switches are used for redundancy, but single-switch implementations are also supported. The following figure shows typical network connectivity.
There are four types of network traffic in a VxRail cluster:
The virtualization and software layers of a VxRail system consist of VMware vSphere and vSAN.
vSphere is the server virtualization product from VMware. It consists of the following main components:
VxRail Manager is a plug-in for VMware vCenter. It provides the software stack for managing compute, storage, and networking in a VxRail environment. It is an extension of VMware vCenter that provides information, upgrades automation, adds or removes nodes, and manages drive replacement. VxRail Manager is shown in the following figure.
VxRail Manager includes the following functionalities:
To view the VxRail Dashboard, select VxRail from the vSphere Client menu. Features include:
To view VxRail monitoring, select a cluster, then select Monitor > VxRail. Features include:
To view the VxRail configuration, select a cluster, then select Configure > VxRail. Features include:
The following actions are available in the cluster and hosts context menu:
vSAN is integrated into the kernel of vSphere and provides the software-defined storage layer. A vSphere cluster is a collection of ESXi hosts that is configured to share resources. vSAN creates distributed, shared storage by aggregating locally attached disks from the hosts that form the vSphere cluster. The following figure shows the vSAN high-level architecture:
VMware Site Recovery Manager (SRM) is deployed in a paired configuration that includes a protected site and a recovery site. The SRM software appliance is deployed at both sites. A vCenter Server instance is required at both the protected and recovery site.
Requirements for SRM installation are:
The following figure shows basic SRM deployment:
A virtual RecoverPoint Appliance (vRPA) is installed in each vSphere cluster environment. The appliance manages all aspects of the data replication process. It uses IP protocol to communicate with the vSphere datastore. The vSphere Web Client is used to protect and recover VMs to any point in time with integrated orchestration and automation capabilities.
The Recover Point for VMs (RP4VMs) system consists of the following components:
The following figure shows the RP4VMs system architecture with the vSphere HTML5 plug-in:
(Source: Dell EMC Integrated Data Protections Product Guide)
VMware vSphere Replication provides asynchronous replication for VMs. It comes fully integrated with vCenter Server and the vSphere Web Client. vSphere Replication provides local data protection, but it can also be used for disaster recovery between sites if some data loss can be tolerated. It can be combined with VMware Site Recovery Manager (SRM) to provide automated failover and failback to minimize downtime.
Industrial gateways, deployed in Purdue model levels 0 through 2 of a factory, allow various factory devices and industrial protocols to communicate with the rest of the network. This communication is accomplished with software that interfaces with protocols including, but not limited to, Modbus, OPC UA, EtherNet/IP, and Profibus. The gateway hosting this software is configured to send factory edge data northbound for tasks such as data analytics. Because industrial gateways differ from standard PCs or servers, they must often be ruggedized to handle harsh environments.
Sensors are an integral source for edge data. Sensors measure a physical property such as vibration, level of light, motion, temperature, and humidity. Sensors are widely deployed in factory environments to ensure that the manufacturing process is properly working or to ensure that the environment is operating at an expected norm. With the current evolution of Artificial Intelligence (AI) and Machine Learning (ML), it is beneficial to obtain, aggregate, and then analyze this sensor data to provide insights into topics like predictive maintenance or process improvement.
The PTC application stack that is detailed in the following table runs on various Dell servers and operating systems running Windows Server or Linux. It is supported by Apache web services and a database repository.
Software component | Type/version | Product version |
Operating systems | Windows Server 2016, 2019 | PTC Kepware Server EX v6.10 |
Red Hat 8.x | PTC ThingWorx v9.1/9.2/9.3 | |
Ubuntu 18.04 LTS | PTC Kepware Edge v1.4 | |
Windows Server 2016, 2019 | PTC ThingWorx v9.1/9.2/9.3 | |
Windows Server 2016, 2019 | Digital Performance Management (DPM) 1.1 | |
Web service | Apache tomcat | Tomcat version 9.3 |
Database | PostgreSQL | PostgreSQL v 13.1 |
Microsoft SQL Server | Microsoft SQL Server 2019 | |
InfluxDB | InfluxDB v1.8.9 |
Other products are available for PTC HA, distributed database, and load balancing. They include PTC connection servers, PTC ThingWorx HA Cluster Service, Apache ZooKeeper, Apache Ignite, PostgreSQL Pool, and others. They are not included in the PTC Edge Solution with Dell EMC VxRail, but can be deployed if you choose.
PTC applications and VxRail offer an extensive set of configuration tools for alerts and monitoring. Various log levels can be configured for tracing and events. Simple Network Management Protocol (SNMP) traps can be configured to integrate with other network monitoring tools like Dell OpenManage.
PTC ThingWorx offers configuration of notifications for specific ThingWorx projects and tags and allows configuration of Mashups for reporting and tracking. Various triggers can be configured using specific sources, event handlers, and recipients for timely alerts and handling of key events.
PTC Kepware Server integrates with Windows event log to report application and system events. Alarms and events can be configured along with an SNMP agent for sending notifications.
Dell EMC VxRail Manager offers provisioning, management, and performance monitoring along with a rich set of integrations with REST API, SNMP, and others.
The ThingWorx Industrial IoT Solutions Platform offers a rich set of capabilities that enable powerful solutions for design, manufacturing, service, and industrial operations.
Flexible and purpose-built for IIoT, ThingWorx empowers industrial enterprises to digitally transform every aspect of their business with innovative solutions that are simple to create, easy to implement, and scalable to meet future needs and accelerate time to value.
ThingWorx solves manufacturing challenges with prebuilt options that simplify industrial connectivity. With a library of client drivers, advanced plug-ins, and connectors, ThingWorx enables connectivity to thousands of devices and systems, establishing a single source for collecting, aggregating, and enabling secure access to IIoT data. The following figure shows how ThingWorx Foundation enables different operations in a typical manufacturing environment.
Digital Performance Management (DPM) is a closed-loop, problem-solving solution, as shown in the following figure. It helps manufacturers identify, prioritize, and solve their biggest loss challenges. This data results in reduced cost, increased revenue, and improved service levels.
Kepware Server (KEPServerEX) is software designed for accurate communications, quick setup, and interoperability between client applications, industrial devices, and systems. The server provides a wide range of plug-ins, device drivers, and components that suit most communication needs. The plug-in design and single-user interface provide consistent access from standards-based applications and nonstandards-based applications with native interfaces. KEPServerEX supports the OPC Unified Architecture (OPC UA) specification and many of the OPC Classic specifications, including OPC Data Access (OPC DA), OPC Alarms and Events (OPC AE), and OPC Historical Data Access (OPC HDA).
KEPServerEX can seamlessly stream real-time industrial control data directly into big data and analytic software for Business Intelligence (BI) and Operational Excellence (OE). Its customizable data format supports most MQTT and REST applications—enabling users to choose the vendors and communication methodologies that are ideal for their system. The KEPServerEX connectivity platform offers simultaneous accessibility to client applications through a list of supported industry standards, proprietary technologies, and native client interfaces such as MQTT, ODBC, SNMP, and REST. The following figure shows KEPServerEX interaction with different interfaces and drivers.
(Source: KEPServerEX Product Page)
ThingWorx Kepware Edge allows the most valuable features of KEPServerEX to be deployed in Linux-based environments, enabling connectivity directly at the site of the machine, device, or sensor. These features enable an improved connectivity architecture when connecting automation applications (like SCADA, MES, or Historians) or IoT applications to remote PLCs, sensors, or other devices. The product can connect to:
The product provides data access from these devices to:
OPC is the interoperability standard for the secure and reliable exchange of data in the industrial automation space and in other industries. It is platform-independent and ensures the seamless flow of information among devices from multiple vendors. The OPC Foundation is responsible for the development and maintenance of this standard. The introduction of service-oriented architectures (SOA) in manufacturing systems introduced new challenges in security and data modeling. The OPC Foundation developed the OPC Unified Architecture (OPC UA) specifications to address these issues while simultaneously providing a feature-rich technology and open-platform architecture that is future-proof, scalable, and extensible.
OPC UA goes beyond the limitations of existing OPC interfaces and provides solutions for numerous new use cases. The functionality and enhancements of OPC UA allow for implementing and using solutions on different platforms. The OPC UA standard is a platform-independent communication framework that is rapidly emerging across all types of industry.
Security is essential when exchanging information between systems. A sophisticated security-in-depth concept is built into the core of OPC UA. OPC UA defines application-level and transport-level security, built into layers on top of the TCP/IP stack. OPC UA introduces the UA-Secure-Conversation layer in addition to the Secure Sockets Layer (SSL) and Transport Layer Security (TLS) used in HTTPS standards of modern web applications.
For standard use cases, Unified Automation developed standard products using the advantages of OPC UA. These products can be used out-of-the-box and can be combined with other systems supporting OPC UA or legacy OPC standards.
As a standard for data interoperability, the OPC Foundation facilitates the adoption of OPC technology through collaboration with other international standards organizations. These collaborative efforts provide members with new opportunities for market expansion and technology integration, as the following figure shows.
The Modbus communication protocol is the oldest and most popular automation protocol in the field of process automation and Supervisory Control and Data Acquisition (SCADA). Modbus provides a common language for devices and equipment to communicate with one another.
For example, Modbus enables measurement devices that are connected on the same network to communicate the results to a supervisory computer or programmable logic controller (PLC). The Modbus Organization is an association of users and suppliers of Modbus-compliant devices, and they manage the development and updates of Modbus protocols.
Several versions of the Modbus protocol exist for the serial port and Ethernet. The most common are:
Modbus is an open standard that describes the messaging communication dialog. Modbus communicates over several types of physical media such as:
PROFINET is a mechanism to exchange data between controllers and devices. Controllers could be PLCs, Distributed Control Systems (DCSs), or Programmable Automation Controllers (PACs). Devices could be I/O blocks, vision systems, RFID readers, drives, process instruments, proxies, or even other controllers.
PROFINET, the leading industrial Ethernet standard for automation, exchanges data quickly and deterministically. It creates integrated processes and configures networked interfaces. The openness and flexibility of PROFINET ensures freedom when designing a machine and system structure. Designed for efficiency, it optimizes use of available resources and significantly increases plant availability.