Do you need software-defined storage (SDS) for your enterprise virtualization, analytics, and enterprise sync and share workloads? If so, then Red Hat has the program for you: Red Hat Gluster Storage 3.2.
Red Hat likes to say its open-source Gluster SDS Scale-out network-attached storage (NAS) is for the modern enterprise workload. It’s not wrong.
You can deploy Gluster quickly and use it to flexibly manage your ever-increasing storage needs without tears. The program is a combination of Red Hat Enterprise Linux (RHEL) and GlusterFS. GlusterFS is an open-source scalable network file-system. You can use it with your existing storage infrastructure and easily add common off-the-shelf (COTS) storage as your data-storage requirements expand.
Red Hat claims this newest edition includes a number of enhancements and new features. Its goal is to improve small-file performance, provide data integrity at a lower cost, and enhance Red Hat OpenShift Container Platform and container integration in general.
In particular, Red Hat Gluster Storage 3.2 introduces the following capabilities and improvements:
Deeper integration with Red Hat OpenShift Container Platform: It does this by adding native support for advanced storage services such as geo-replication and in-flight encryption for applications deployed in containers. These enhancements, packaged into the refreshed docker container image that is shipped with the latest release of the product, can also enable three times as many persistent volumes (PVs) per cluster according to Red Hat quality engineering testing.
Smaller storage hardware footprint: The program accomplishes this by using arbiter volumes. Instead of holding replicated data, these only contain file/directory names (i.e. the tree structure) and extended metadata attributes. This gives you more room for data, thus shrinking your infrastructure costs while maintaining high data integrity. Arbiter volumes also can resolve conflicts in the event of data mismatch between two nodes without requiring a third copy of the data.