IBM says its new all-flash systems provide three times the storage capacity in the same space as the current FlashSystem 900 and reduce data capacity costs by 60%. Credit: Reuters/Lucy Nicholson All-flash storage has become increasingly popular in data centers as a means of much faster data access than traditional hard disk, but its growth has been impeded by cost and storage density. There was too much of the former and too little of the latter. Every memory, storage and server vendor is working full out to address that issue, and it has turned into quite an arms race, which benefits the customer. So much so that Gartner predicts that within the next 12 months, solid-state arrays will improve in performance by a factor of 10 while doubling in density and cost-effectiveness. IBM has just made its contribution to that growth. It has announced advances in flash storage that it claims will provide a three-fold increase in density in the same physical space for its FlashSystem 900 flash arrays, while reducing data capacity costs by 60 percent. IBM used Micron’s newest 3D stacking memory to achieve this density. A few years back, NAND flash memory hit a wall at around 16nm, where it became impossible to shrink the memory cells any smaller. That stopped the growth of memory density and thus capacity when they couldn’t squeeze any more bits into the same space. The solution that flash memory makers like Micron came up with was to stack the layers of flash memory on top of each other like pancakes. If they couldn’t scale out, they could scale up. Very quickly they expanded from 32 layers to 48 layers to 64 layers, and Western Digital is talking about 96-layer NAND flash sometime in the near future. With the stacking comes increased capacity again. That’s how IBM is able to get three times the storage capacity in the same space. The current FlashSystem 900 on the market – which has been on the market since April of last year, so it was due for an upgrade – has 60TB of capacity. Now it has a max capacity of 180TB. IBM’s secret sauce It’s the same trick everyone uses, since Micron has many partners, but IBM has its own secret sauce, as well. It uses hardware compression with a custom FPGA, so the array doesn’t need to talk to the CPU to perform the compression. It’s all done in the flash array itself. This allows for much faster speed. IBM also claims there is no single point of failure in the array for always-on performance. New IBM virtualization software At the same time, IBM also introduced new virtualization software called Spectrum Virtualize, which is designed to ease the migration of data to and from its public cloud services — for simple data migrations, as well as for disaster recovery of data. Finally, new software enables IBM and non-IBM storage to be used with the Docker and Kubernetes containers environments, and new cloud-based software integrates storage with artificial intelligence and machine learning to provide constant diagnostics to help optimize the performance, capacity and health of clients’ storage infrastructure. IBM said the new all-flash systems and new software would be available by the end of the year. Related content news analysis AMD launches Instinct AI accelerator to compete with Nvidia AMD enters the AI acceleration game with broad industry support. First shipping product is the Dell PowerEdge XE9680 with AMD Instinct MI300X. By Andy Patrizio Dec 07, 2023 6 mins CPUs and Processors Generative AI Data Center news analysis Western Digital keeps HDDs relevant with major capacity boost Western Digital and rival Seagate are finding new ways to pack data onto disk platters, keeping them relevant in the age of solid-state drives (SSD). By Andy Patrizio Dec 06, 2023 4 mins Enterprise Storage Data Center news Omdia: AI boosts server spending but unit sales still plunge A rush to build AI capacity using expensive coprocessors is jacking up the prices of servers, says research firm Omdia. By Andy Patrizio Dec 04, 2023 4 mins CPUs and Processors Generative AI Data Center news AWS and Nvidia partner on Project Ceiba, a GPU-powered AI supercomputer The companies are extending their AI partnership, and one key initiative is a supercomputer that will be integrated with AWS services and used by Nvidia’s own R&D teams. By Andy Patrizio Nov 30, 2023 3 mins CPUs and Processors Generative AI Supercomputers Podcasts Videos Resources Events NEWSLETTERS Newsletter Promo Module Test Description for newsletter promo module. Please enter a valid email address Subscribe