After many months of infrastructure re-architecting, KMi Systems have deployed the University’s first hyperconverged platform for its VMware vSphere estate.
Hyperconvergence is a type of infrastructure system with a software-centric architecture that tightly integrates compute, storage, networking and virtualization resources and other technologies from scratch in a commodity hardware box supported by a single vendor.
Maxta, the vendor of KMi’s new hyperconverged layer within its virtualised estate, takes a different stance on hypervisor storage management when compared to traditional Storage Area Network (SAN) / Direct Attached Storage (DAS) solutions. These require expensive disk arrays or ‘Just a bunch of disks’ (JBODs) with their disks configured with a mixture of different ‘Random Array of Inexpensive Disks’ (RAID) sets and incurs a high amount of administration as a result. RAID sets have often restricted us from providing small increments to storage capacity owing to the nature of the technology, therefore making storage expansion a costly exercise. However, KMi’s new hyperconverged layer in its virtual estate has changed this.
Maxta maps raw disk to the hypervisor, aggregating it into a large pool. Redundancy is then leveraged via data replication to multiple nodes, akin to Hadoop Distributed File System (HDFS). This approach allows for cheaper storage increments, whilst still affording us cross-datacentre redundancy.
Unlike our previous SAN and current DAS solution, we are also reaping the rewards of data compression on the backend disk, therefore allowing us to fit more data into our pool of aggregated storage.
A further benefit from the Maxta technology is its proprietary snapshotting. This is particularly useful for KMi's Data Science / Big Data research allowing snapshots with zero interruption to the running services. Traditionally, snapshotting technology built into virtualisation solutions institutes a minor interruption to service when creating a snapshot this has a negative impact on running big data tools. The Maxta solution and its new approach mitigates this issue.
This emerging technology is cutting edge, and will eventually be used as a staple requirement for virtualisation in enterprise datacenters across the world. KMi’s early adoption of this is befitting of the department’s ethos, and further supplements our researchers and developers server environment, whilst also leading the way for the University in this field.
For more information about Maxta and the features it affords KMi, see their website for more details at www.maxta.com
