Virtualization and blade server technologies have enabled a generation of consolidated computing devices capable of cramming extraordinary computing power into smaller form factors. But the increased processing power per square inch has brought about a new I/O problem: The pipes can't move data fast enough to keep up with today's processors. To address that problem, new I/O virtualization products and standards are emerging to extend PCI Express (PCIe) pathways to separate I/O devices. This allows multiple physical servers and virtual machines (VMs) to share I/O resources.
Industry standard bodies are getting involved as well. The PCI Special Interest Group (PCI-SIG), which handles PCI specifications, has developed and formalized two standards specific to IO virtualization: Single-Root IOV (SR-IOV) and Multi-Root IOV (MR-IOV).
According to the PCI-SIG, SR-IOV allows multiple guest operating systems to simultaneously access an I/O device without requiring a hypervisor on the main data path. MR-IOV builds upon the SR-IOV standard by allowing access to PCI- or SR-IOV-compliant I/O devices over a shared PCIe fabric. The goal of the standards is to enable multiple separate servers to access and share multiple I/O cards inside one or more card cages. Both SR-IOV and MR-IOV meet that goal, but neither standard has seen any significant vendor adoption.
While vendors seem to be eschewing industry standards, that hasn't stopped their progress in the market; they're pushing proprietary solutions and are starting to find market traction and early adopter customers.