Intel Corporation

Suvansh Krishan Kapur

Raman Nayyar

Linda J Rankin

Kai Cheng

Lily Pao Looi

Nitin Y Borkar

D Michael Bell

Ali S Oztaskin

Suresh S Chittor

Daniel A Manseau

Kevin Koschoreck

Srinand Venkatesan

Frank Verhoorn

Linda C Ernst

Robert J Hoogland

Joseph Bonasera

Methods of obtaining, enqueueing and executing several memory transactions are described, where the memory transactions may be generated in a first order but executed in a second order. Despite the relaxed ordering, essential programming paradigms such as producer-consumer relationships are not affected. Chipsets and systems using the methods are also described and claimed.

Method and apparatus for exploiting parallelism across multiple traffic streams through a single channel

An apparatus and method for controlling data traffic flow and data ordering of packet data between one or more peripheral devices and a processor/memory combination by using a packet processing engine, located within an input-output hub (IOH). The IOH may include a packet processing engine, and a switch to route packet data between the one or more peripheral devices and the packet processing engine. The packet processing engine of the IOH may control data traffic flow and data ordering of the packet data to and from the one or more peripheral devices through the switch and also maintains flow and ordering to the processor/memory subsystem. The packet processing engine may be operable to perform packet processing operations, such as virtualization of a peripheral device or Transmission Control Protocol/Internet Protocol (TCP/IP) offload.

Transaction flow and ordering for a packet processing engine, located within an input-output hub

A shared bypass bus structure for low-latency coherency controller access in a coherent scalable switch. In a coherent scalable switch with multiple coherent interconnect ports, distributed coherency control structures, and a crossbar interface between them, a shared bypass bus permits data transfer between the coherent interconnect ports and the coherency control structures while bypassing the crossbar interface. Some embodiments may comprise scalable switches to support one or more sets of processors with substantially independent snoop or cache coherency paths or arrangements.

Shared bypass bus structure

A bus expander bridge is provided for interfacing first and second external buses (such as PCI buses) to a third bus. The bus expander bridge is configurable in either an independent mode in which the first and second external buses operate independently and a combined mode in which the first and second external buses are combined to create a single bus. The bus expander bridge includes a first set of data queues for routing data between the first external bus and the third bus, and a second set of data queues for routing data between the second external bus and the third bus. The bus expander bridge also includes a controller coupled to the first and second sets of data queues and operating the first and second sets of data queues in parallel for the independent mode. The controller routes even addressed data through the first set of data queues and routes odd addressed data through the second set of data queues for the combined mode.

Dual mode bus bridge for computer system

A method for processing an atomic transaction in a multi-bus system that includes a local bus and a remote bus. According to the method, a first transaction in an atomic sequence is received from the local bus. The first transaction is terminated on the local bus. The first transaction is performed on the remote bus. A response to the first transaction is received and the response is placed on the local bus.

Bus protocol for atomic transactions

A massively parallel data processing system is disclosed. This data processing system includes a plurality of nodes, with each node having at least one processor, a memory for storing data, a processor bus that couples the processor to the memory, and a remote memory access controller coupled to the processor bus. The remote memory access controller detects and queues processor requests for remote memory, processes and packages the processor requests into request packets, forwards the request packets to the network through a router that corresponds to that node, receives and queues request packets received from the network, recovers the memory request from the request packet, manipulates local memory in accordance with the request, generates an appropriate response packet acceptable to the network and forwards the response packet to the requesting node.

Growing community of inventors

Portland, OR, United States of America

Suvansh Krishan Kapur