As hyperscale service providers overhaul their data center gear with the latest technology, they're selling off their old equipment, which is likely more powerful than most enterprises use today. Credit: Monsitj / Getty Images A company that specializes in creating second lives for IT hardware is expanding its initiative to reengineer and sell decommissioned data-center equipment from the major hyperscale operators that are aggressively replacing relatively new hardware. ITRenew announced the plan at the recent Open Compute Project (OCP) conference, promising to sell full servers previously owned by the big operators, reengineered, warrantied, and configured for turnkey uses like web serving and Kubernetes. ITRenew launched its first server racks two years ago and is now making the initiative more broadly available to all industries so more potential customers can buy OCP-certified hardware. OCP’s CTO Bill Carter said there’s a proliferation of retired gear because of the massive growth of hyperscalers that would rather put newer, more powerful gear in the same space than build out new space. With the core counts from Intel and AMD skyrocketing, a 2U server from 2020 is much more powerful than a 2U from 2018. And Carter noted, “that’s newer than a lot of equipment being used in IT these days.” He said reselling whole servers wasn’t done even a year ago; recycled gear was sold on a component level. “When you have these mega data centers commissioning 100,000 units every year, you’re also taking out tens of thousands of old equipment every quarter. Now I can set up an assembly line to recondition it,” he said. Ali Fenn, president of ITRenew, also noted that hyperscale refresh cycles are very aggressive. Many of these hyperscale operators don’t even have warranties, she said, because they do not need the support. If a server breaks down, there are tens of thousands of fallback servers and they just take it out of service and replace it. The equipment sold by ITRenew and OCP comes from a number of original design manufacturers (ODM) like WiWynn and Quanta that sell unbranded equipment to the hyperscalers. ITRenew turns it into rack-scale gear for the broader global markets and brands them as Sesame by ITRenew. It does not use gear from the big OEMs like HPE and Dell. ITRenew integrates components needed for telcos, service providers and mainstream enterprises, refreshes and recertifies every device, and provides a three-year standard warranty with an option for six years. “This is advanced, high-performance, proven technology, and the failure rates we see are on par with new equipment. Consider this the equivalent of certified, pre-owned BMWs with five thousand miles on them,” she said via email. ITRenew notes that hyperscalers turn over their technology after just one or two years. The tier 2/3/4 data centers that buy the rebranded gear get access to premium technology at cost-effective price points. Price depends on configuration, but ITRenew generally says its total cost of ownership is 40-50% lower than for comparable OEM equipment. The ITRenew solutions arrive on site fully assembled and ready to plug in. Fenn said ITRenew customers are primarily tier 2/3/4 services providers, enterprises doing private cloud, and telcos, globally. Most common workloads include a variety of web apps, Kubernetes, virtualized environments, converged, and AI/ML. “Customers are those that have wanted access to ODM equipment but who have as yet been unable to get due to the inability of the major ODMs to deliver full solutions, and to service and support smaller scale customers than the hyperscalers,” she said. Many off-brand ODMs like Inspur and WiWynn sell only to the tier 1 hyperscalers (such as Amazon, Facebook and Google) but have not made an effort to build up sales teams to sell to the enterprise. After all, when you have a half-dozen customers buying a quarter-million servers every month, why bother? Sesame servers are sold by ITRenew and its partners and featured on the OCP Marketplace website. Related content news High-bandwidth memory nearly sold out until 2026 While it might be tempting to blame Nvidia for the shortage of HBM, it’s not alone in driving high-performance computing and demand for the memory HPC requires. By Andy Patrizio May 13, 2024 3 mins CPUs and Processors High-Performance Computing Data Center news CHIPS Act to fund $285 million for semiconductor digital twins Plans call for building an institute to develop digital twins for semiconductor manufacturing and share resources among chip developers. By Andy Patrizio May 10, 2024 3 mins CPUs and Processors Data Center news HPE launches storage system for HPC and AI clusters The HPE Cray Storage Systems C500 is tuned to avoid I/O bottlenecks and offers a lower entry price than Cray systems designed for top supercomputers. By Andy Patrizio May 07, 2024 3 mins Supercomputers Enterprise Storage Data Center news Lenovo ships all-AMD AI systems New systems are designed to support generative AI and on-prem Azure. By Andy Patrizio Apr 30, 2024 3 mins CPUs and Processors Data Center PODCASTS VIDEOS RESOURCES EVENTS NEWSLETTERS Newsletter Promo Module Test Description for newsletter promo module. Please enter a valid email address Subscribe