High performance memory based communications interface
AssigneeLiquid Computing Corporation
InventorsKaiyuan HUANG, Michael F. KEMP
Life of this patent
Estimating the term…
Abstract
Embodiments of the present invention include enhanced functionalities and components within a Communication Endpoint Processor (CEP) that act as an interface between computational and communications domains. The embodiments disclosed herein deliver a complete memory mapped high performance interface that has the ability to support the simultaneous transmission of multiple frames of multiple sizes, and that has the ability to interrupt the transmission of lower priority frames in order to send higher priority frames.
Claims21
A method for a local computational host to communicate with a remote computational host, the local and remote computational hosts each comprising an operating system, a workload;
- a communication endpoint processor (CEP) kernel coupled to the workload over an interface and a CEP engine coupled to the CEP kernel over a CEP engine interface, the method comprising the steps of:
- the local workload invoking a one sided remote memory operation across the interface to the local CEP kernel;
- the local CEP kernel constructing a memory operation command;
- the local CEP kernel sending the constructed memory operation command across a local CEP kernel to remote CEP engine protocol;
- the remote CEP engine consuming and executing the sent memory operation command without invoking services of the remote CEP kernel, the remote workload or of the remote operating system, and
- the local CEP engine receiving one of a result of the executed memory operation command and a confirmation that the sent memory operation command has been carried out.
A method for optimizing a data packet in a high performance computer system, comprising the steps of:
- preparing a payload of the data packet;
- passing the prepared payload to a communications endpoint processor (CEP) kernel;
- appending, by the CEP kernel, a header to the prepared payload to generate a CEP kernel protocol unit that is optimized for a processor environment;
- transmitting, by the CEP kernel, the CEP kernel protocol unit to a CEP driver that is optimized for high throughput and low latency, and
- transforming, by the CEP driver, the transmitted CEP kernel protocol unit into one of a plurality of CEP kernel protocol unit formats depending upon a size range of the payload, each of the plurality having a different format for efficient handling by a CEP engine that is configured to automate communication tasks and to interface to a system memory controller for direct memory operations.
Dependencies are inferred from the claim text.