When it comes to data storage, there is always one basic concern. No matter how much data has to be accessed, the decision concerning what storage platform to use usually comes down to how fast that data needs to be accessed. Increasingly, large amounts of data now need to be accessed faster than ever.
To address that issue, Exablox today unveiled an all-Flash configuration of a storage array based on a distributed object system. Sean Derrington, senior director of products for Exablox, says that, rather than being relegated to secondary storage, distributed object systems are now being increasingly used for primary storage in scenarios where the amount of data that needs to managed exceeds what can be handled by traditional file system.https://o1.qnsr.com/log/p.gif?;n=203;c=204663295;s=11915;x=7936;f=201904081034270;u=j;z=TIMESTAMP;a=20410779;e=iThe OneBlox 5210 makes it possible for IT organizations to support that use case via an array that allows them to plug their choice of solid-state drives into an Exablox array. That approach, says Derrington, effectively drives the cost of providing Flash storage using the Exablox array down to 50 cents per GB because Exablox doesn’t require customers to buy SSDs from them.
Derrington adds that IT organizations can upgrade the OneBlox 5210 any time they want by plugging in faster SSDs as they become available. In contrast, other storage vendors make customers wait six months or more for them to validate a new SSD that they usually mark up to increase SSD costs even more, says Derrington.
Finally, Derrington notes that the distributed object system created by Exablox provides IT organizations with more granular control over the size of objects being accessed by any given application which, in turn, improves overall performance. Exablox claims it can achieve 50,000 IOPS per node while scaling to more than 700 TB per cluster. A total of 1.4 petabytes of storage that can be accessed by both modern and legacy applications compatible with NFS and SMB protocols can be packed into each array.
There is clearly no shortage of options these days when it comes to data storage. Most IT organizations are going to deploy different types of systems to support a multitude of application workload scenarios. Having more flexibility in terms of supporting applications like Big Data that are only going to continue to grow over time is always a good thing.