5 Things to Know about PureApplication System V2 support for HADR
5GS6_Margaret_Ticknor 1000005GS6 Visits (8567)
IBM PureApplication System provides many capabilities that allow you to implement high availability strategies (HA) and disaster recovery scenarios (DR) for your workloads. You might consider PureApplication System as a toolbox providing you the tools to implement HA strategies and DR scenarios.
Here are 5 Things to Know about PureApplication System support for HADR, for more detailed information grab a copy of SG24-8246-00 Impl
IBM General Parallel File System (GPFS) is a scalable, high performance file management infrastructure for IBM AIX®, Linux, and Windows Server systems based on a shared disk model, provided by an underlying SAN. GPFS provides fast, reliable access to data from multiple nodes in a cluster environment. Applications can readily access files using standard file system interfaces, and the same file can be accessed concurrently from multiple nodes. GPFS is designed to provide high availability through:
In terms of PureApplication System, GPFS allows multiple nodes or virtual machines to concurrently access the same data. GPFS provides a shared file system to support highly available configurations for virtual system and virtual application pattern deployments within the same rack or across racks.
In order for workloads to connect to a GPFS server and use shared file systems, the shared service for GPFS must be deployed. The shared service defines the parameters to connect to the GPFS server that connects to shared storage. As a result, any pattern deployed to a cloud group into which a shared service is deployed, can utilize the storage.
With PureApplication System V2.0, the GPFS Client Policy supports application components in both virtual system patterns and virtual application patterns. When the GPFS Client Policy is added to a pattern component, at deployment time the GPFS product is installed on the virtual machine. As well, the configuration is retrieved from the shared service, and the client is connected to shared file systems that are hosted by the deployed GPFS server pattern.
A new type of storage volume, called block storage, was introduced with PureApplication System V2.0. Block storage can be cloned on a single rack or replicated to another rack for HADR solutions. Block storage maximizes storage controller LUNs, thus directly avoiding VMFS and allowing more capabilities. Volumes can be internal volumes or defined in an external storage device.
Block storage replication is a form of disaster recovery at the individual storage volume level. Rather than replicating the entire system configuration, you can configure replication of each storage volume between local and remote systems in either direction. Block storage replication removes the need to have a dedicated disaster recovery rack.
A multisystem environment consists of two or more systems connected to each other physically (by networking) and logically (in a defined relationship). The systems in a multisystem environment might be at the same or different geographic locations.
PureApplication System provides the ability for continuous availability for key applications by deploying across multiple systems in a multisystem environment. The multisystem environment contains management domains and deployment subdomains. Systems that are part of a multisystem environment are also referred to as locations.
You can use a multisystem environment to perform a number of tasks, including but not limited to:
Addison Goering is a Certified IT Specialist with the WebSphere Education team. His main specialty is the design, development, and delivery of courses in the WebSphere product family. He has developed and delivered courses ranging from webinars to week-long workshops on products such as WebSphere ESB, IBM Workload Deployer, WebSphere Application Server, WebSphere Business Services Fabric, and WebSphere BPM. He is the lead developer on the WebSphere Education team that is developing education on IBM PureApplication System. Addison holds a B.S. in education from Keene State College in New Hampshire, mainframe certification from DePaul University in Chicago, and several certifications from IBM.