The folks over at Gluster dropped in the news that they are releasing their Gluster Storage platform:
“By integrating the GlusterFS file system with an operating system layer and improved management user interface, Gluster provides a complete storage software platform that simplifies the task of deploying petabyte-scale storage to two installation steps and a few mouse clicks. The new platform addresses the complexity of managing ‘big data,’ the demanding scalability requirements of modern applications”
Now what has this got to do with Virtualization you might ask,
the accelerating growth of Virtual Machine deployments requires people to rething their storage strategy, the demand for a unified storage space that contains high available copies of the running virtual machines is growing.
The Gluster Storage platform comes as a Clustered Storage on s stick version, the complete platform image fits on a USB memory stick that can configure a bare server to a clustered storage node in 15 minutes or less
As mentionned it provides the user with a Unified global namespace, according to them you can add hundreds of petabytes in a single volume across multiple commodity storage nodes.
It has Web-based installation and management providing you with rapid installation of first storage node with disk formatting performed post-install; integrated management of volumes, data resources and servers plus centralized logging and reporting.
And off course it is High availability – data can be replicated (mirrored) for high availability; real time self-healing performs error detection and correction within files while they are running and during recovery from hardware failures
“Gluster Storage Platform is the natural evolution of GlusterFS, combining the power and scalability of clustered storage with unparalleled simplicity,” said Anand Babu (AB) Periasamy, co-founder and CTO of Gluster. “Gluster is the only commercial scale-out storage platform that eliminates metadata server issues and allows non-experts to deploy petabyte-scale storage with the ease of an install wizard.”
With regard to virtualization ,
Gluster Storage Platform ensures continuous operation of virtual machines (VMs). Replicated virtual machines can continuously operate in the event of hardware failure and recovery is performed in the background without requiring a restart or blocking I/O to the live VM. Gluster Storage Platform uses checksum based healing which detects and corrects errors within the VM rather than for the entire VM image.
Gluster argues that as the Gluster Storage Platform aggregates disk and memory resources into a single pool of capacity under a global namespace it provides a true virtual storage environment to complement server virtualization. VM images and application data can be stored in the same system in a single volume, eliminating silos of data and centralizing management. Multiple storage building blocks are clustered together in parallel, eliminating I/O bottlenecks and hotspots that often negatively impact VM performance. With Gluster, storage administrators no longer need to manually manage multiple volumes and individual storage connections.
According to them it is now common for mission critical applications to be deployed on VMs. The VM images not only need to be stored, they also need to continue operating in the event of hardware failures and not allow faults to interrupt applications and services. Gluster Storage Platform employs file replication to ensure multiple copies of a VM are available in the event of hardware failure and provides both high availability and sophisticated real time self-healing to ensure continuous VM operation.
We quickly looked into the GlusterFS platform and set it up on different nodes,
the setup speed and ease of confiugration was a welcome suprise as opposed to different other clustering setups.