IBM expanded its flash array portfolio with DeepFlash 150 for big data workloads.
The DeepFlash 150 offers a high storage density of up to 170TB per rack which means that seven petabytes of all-flash can be placed into one standard rack.
Customers will be able to use the JBOF in a 3U chassis that customers can use as a standalone or with the company’s Spectrum Scale storage software. Spectrum Scale is based on IBM’s General Parallel File System and is capable of running on separate server hardware.
The company said that DeepFlash 150 is designed for companies in need of high capacity, high density and cost-optimised all-flash arrays for big data workloads.
Some of the benefits are that it provides a building block for software-defined storage infrastructures designed for big data, media and entertainment streaming, high-speed database, and hyper-scale environments.
One of the big selling points of the product is its cost, which has long been an issue for all-flash arrays. The company said that the basic hardware platform is priced to under $1 a gigabyte. The hope is that the barrier to entry has been lowered enough by the price that more people will see it as a viable option for their big data workloads.
By combining the all-flash system with Spectrum Scale it becomes more of an array in terms of storage functions, this means that it will then offer snapshots, replication, and compression, encryption is also included in the software layer.
The all-flash market has become a lot more competitive over the past year as significant advances have made the product much cheaper and therefore easier to sell.
Pure Storage is one of the vendors pushing the technology use with unstructured data with FlashBlade and EMC with Project Nitro, an all-flash version of Isilon storage, although this is only just heading into beta stage.
DeepFlash 150 is equipped with 64 multilevel cell flash cards instead of using SAS or SATA solid-state drives. These are provided by Western Digital’s SanDisk division by virtue of a partnership with IBM.
One of the expected use cases for the all-flash system is in-memory analytics, such as SAP HANA, this is because in this use case the system requires the capability to quickly load and read large amounts of data.