ProOnE: a general-purpose protocol onload engine for multi- and many-core architectures

Abstract Modern high-end computing systems utilize spe-cialized offload engines to enhance various aspects of their processing. For example, high-speed networks such as In-finiBand, Quadrics and Myrinet utilize specialized hard-ware to offload network processing to help improve per-formance. However, such hardware units are expensive, and their manufacturing complexity increases exponentially de-pending on the number and complexity of tasks they off-load. On the other hand, the proliferation of multi- and many-core processors into the general desktop and laptop markets is increasingly driving their cost down due to the economies of scale. To take advantage of the obvious ben-efits of multi/many-core architectures, we propose, design and evaluate ProOnE, a general purpose Protocol Onload Engine. ProOnE utilizes a small subset of the available cores on a multi-core CPU to “onload ” various tasks in a ded-icated manner instead of “offloading ” them to specialized hardware. The general purpose processing capabilities of multi-core architectures allow ProOnE to be designed in a flexible, extensible and scalable manner, while benefitin