Intel Corporation

Scott R Tetrick

Glenn J Hinton

Sanjeev N Trika

Bruce A Young

Frank T Hady

Amber D Huffman

Dale J Juenemann

Byron R Gillespie

James A Boyd

Thomas J Barnes

Oscar P Pinto

Scott Burridge

Raghu Murthi

A method and system to perform caching based at least on one or more file-level heuristics. The caching of a storage medium in a caching device is performed by a cache policy engine. The cache policy engine receives file-level information of input/output access of data of the storage medium and caches or evicts the data of the storage medium in the caching device based on the received file-level information. By utilizing information about the files and file operations associated with the disk sectors or logical block addresses of the storage medium, the cache policy engine can make a better decision on the data selection of the storage medium to be cached in or evicted from the caching device in one embodiment of the invention. Higher cache hit rates can be achieved and the performance of the system utilizing the cache policy engine is improved.

Method and system to perform caching based on file-level heuristics

An element of a multi-functional device that integrates a high performance processor into a PCI to PCI bus bridge (P2P). The invention is part of a design that consolidates a high performance processor (the local processor) and other processing elements into a single system which utilizes a local memory. Four PCI interrupt inputs are provided which can be routed to either local processor interrupt inputs or to PCI Interrupt output pins. In this manner, a server designer is able to connect the PCI interrupts directly to the local processor without any jumpers to provide configuration. Additionally, by providing software which would execute on the local processor, the local processor system can intercept the PCI interrupts and process the low level interrupts to create an intelligent I/O subsystem. A simple multiplexor is used to direct the PCI interrupts inputs to the local processor or directs the PCI interrupt inputs directly to the PCI interrupt outputs. The PCI interrupt inputs would be interrupts from PCI devices connected to the secondary PCI bus or PCI add-in cards connected to the secondary PCI bus. The PCI outputs would go directly to an interrupt controller which supports the host processor interrupt structure. This PCI interrupt output mechanism supports the ability to have the local processor intercept the PCI interrupts, determine if the local processor should process the interrupt or forward the interrupt upstream to the host.

Programmable PCI interrupt routing mechanism

Testing of shared memory (RAM) in a multiple processor computer system is achieved by partitioning the memory and allocating the memory portions to respective processors in the system. Each processor performs testing of its allocated memory portion simultaneously with the other processors in order to reduce the time required to complete the memory test.

Memory testing in a multiple processor computer system

Growing community of inventors

Portland, OR, United States of America

Scott R Tetrick