As the annual Open Compute Project (OCP) Summit approaches, we thought we’d take a minute to discuss the benefits composability brings to hardware standardized on OCP specifications.
OCP defines itself as “a collaborative community focused on redesigning hardware technology to efficiently support the growing demands on compute infrastructure.” Essentially the organization creates specifications for hardware designed for wide-scale, disaggregated interchangeability.
Liqid supports the spirit of OCP because it mirrors our own. Liqid solutions are optimized from the ground up to provide adaptive, software-defined composability for off-the-shelf hardware. Liqid's disaggregated model for scaling composable is literally out of the box, free from the sales-cycle trap of being forced to level up hardware according to new product schedules — and the tech limitations that go along with that way of building data capability.
One of the core missions of OCP is to address the shortcomings of traditional infrastructure for enterprises and large service providers. As workloads become more dynamic, the infrastructures must also become more dynamic.
Composable is the answer to this challenge, creating efficiency through resource optimization, operational optimization and the least amount of manpower to employ. Liqid composes bare-metal servers from resources housed in industry-standard chassis, and can compose more GPUs than a typical server can support.
Composable disaggregates physical resources and puts storage, networking and compute back into free pools of resources alongside GPU, FPGA, and Intel Optane memory technology. It uses software-defined tech to gather resources and then allocate them as needed across all industry-standard fiber types, including PCIe and Ethernet.