Intel Corporation

Mindspeed Technologies, Inc.

Stmicroelectronics Gmbh

Duane E Galbi

Matthew James Adiletta

Jose Niell

Thomas E Willis

Amit Kumar

Francesc Guim Bernat

Reshma Lal

Yadong Li

Daniel J Lussier

William Duggan

Salma Mirza

Patrick L Connor

Wilson Parkhurst Snyder, Ii

Bill Nale

Susanne M Balle

Alexander Bachmutsky

Aaron Gorius

Hugh Wilkinson

Bradley A Burres

Nagabhushan Chitlur

Joseph B Tompkins

Mrittika Ganguli

Brad Burres

Sundar Nadathur

Bruce Burns

Praveen B Mosur

Andrzej Kuriata

Steven Charles Docker

Ranjit D Loboprabhu

Benjamin J Cahill

Jeongnim Kim

Anandhi Jayakumar

Methods, apparatus, and software for remote storage of hardware microservices hosted on other processing units (XPUs) and SOC-XPU Platforms. The apparatus may be a platform including a System on Chip (SOC) and an XPU, such as a Field Programmable Gate Array (FPGA). Software, via execution on the SOC, enables the platform to pre-provision storage space on a remote storage node and assign the storage space to the platform, wherein the pre-provisioned storage space includes one or more container images to be implemented as one or more hardware (HW) microservice front-ends. The XPU/FPGA is configured to implement one or more accelerator functions used to accelerate HW microservice backend operations that are offloaded from the one or more HW microservice front-ends. The platform is also configured to pre-provision a remote storage volume containing worker node components and access and persistently store worker node components.

Remote storage for hardware microservices hosted on XPUs and SOC-XPU platforms

Examples described herein relate to an apparatus that includes at least two processing units and a memory hub coupled to the at least two processing units. In some examples, the memory hub includes a home agent. In some examples, the memory hub is to perform a memory access request involving a memory device, a first processing unit among the at least two processing units is to send the memory access request to the memory hub. In some examples, the first processing unit is to offload at least some but not all home agent operations to the home agent of the memory hub. In some examples, the first processing unit comprises a second home agent and wherein the second home agent is to perform the at least some but not all home agent operations before the offload of at least some but not all home agent operations to the home agent of the memory hub. In some examples, based on provision of the at least some but not all home agent operations to be performed by the second home agent, the second home agent is to perform the at least some but not all home agent operations.

Memory hub providing cache coherency protocol system method for multiple processor sockets comprising multiple XPUs

An apparatus is described. The apparatus includes a memory controller having logic circuitry to write a unit of write data into a plurality of memory chips according to a striping pattern that includes multiple protected sub words, each protected sub word including a smaller portion of the unit of write data and error correction coding (ECC) information calculated from the smaller portion of the unit of write data.

Memory striping approach that interleaves sub protected data words

An apparatus is described. The apparatus includes logic circuitry to multiplex on a data bus a first data burst, a second data burst, a third data burst and a fourth data burst having different respective base target addresses that respectively target a first memory rank, a second memory rank, a third memory rank and a fourth memory rank. A first data transfer for the first data burst occurs on a first edge of a first pulse of a data strobe signal for the data bus and a second data transfer for the second data burst occurs on a second edge of the first pulse of the data strobe signal. A third data transfer for the third data burst occurs on a first edge of a second pulse of the data strobe signal for the data bus and a fourth data transfer for the fourth data burst occurs on a second edge of the second pulse. The second pulse immediately follows the first pulse on the data strobe signal. The first memory rank, the second memory rank, the third memory rank and the fourth memory rank are on a same memory module.

Synchronous dynamic random access memory (SDRAM) dual in-line memory module (DIMM) having increased per data pin bandwidth

Technologies for flexible I/O protocol acceleration include a computing device having a root complex, a smart endpoint coupled to the root complex, and an offload complex coupled to the smart endpoint. The smart endpoint receives an I/O transaction that originates from the root complex and parses the I/O transaction based on an I/O protocol and identifies an I/O command. The smart endpoint may parse the I/O transaction based on endpoint firmware that may be programmed by the computing device. The smart endpoint accelerates the I/O command and provides a smart context to the offload complex. The smart endpoint may copy the I/O command to memory of the smart endpoint or the offload complex. The smart endpoint may identify protocol data based on the I/O command and copy the protocol data to the memory of the smart endpoint or the offload complex. Other embodiments are described and claimed.

Technologies for flexible protocol acceleration

Technologies for flexible I/O endpoint acceleration include a computing device having a root complex, a soft endpoint coupled to the root complex, and an offload complex coupled to the soft endpoint. The soft endpoint establishes an emulated endpoint hierarchy based on endpoint firmware. The computing device may program the endpoint firmware. The soft endpoint receives an I/O transaction that originates from the root complex and determines whether to process the I/O transaction. The soft endpoint may process the I/O transaction or forward the I/O transaction to the offload complex. The soft endpoint may encapsulate the I/O transaction with metadata and forward the encapsulated transaction to the offload complex. The soft endpoint may store responses from the offload complex in a history buffer and retrieve the responses in response to retried I/O transactions. The I/O transaction may be a PCI Express transaction layer packet. Other embodiments are described and claimed.

Technologies for flexible I/O endpoint acceleration

Approaches to organizing/constructing a register file base cell in a way that reduces the number of signals which need to be routed to and through the bit base cell are disclosed. Base cells so constructed allow industry standard static timing approaches and tools to verify the timing of a register file comprised of such base cells as a whole and allow industry standard place-and-route (APR) tools to be used to implement the connections between the base cells and the other register file logic not directly included in the base cell.

Apparatuses for register file with novel bit cell implementation

In general, in one aspect, the disclosure describes an apparatus for engineering di/dt. The apparatus includes a plurality of functional blocks to perform different functions. The apparatus also includes a clock source to provide a clock signal to said plurality of functional blocks. At least one gating device is used to regulate application of the clock to the plurality of functional blocks. A controller is included to control the at least one gating device and turning-on of the clock signal.

Regulating application of clock to control current rush (DI/DT)

Methods and apparatus for performing error correction code (ECC) coding techniques for high-speed implementations. The ECC code word is structured to facilitate a very fast single-error-detect (SED) that allows state machines to be stopped within a single cycle when an error is detected and enables a corresponding single-error-correct (SEC) operation to be performed over multiple cycles while the state machines are in a suspended mode.

ECC coding for high speed implementation

An integrated circuit for processing communication packets having separate data buffers and separate state information buffers. Each data buffer and each state information buffer (hereinafter termed resources) has an associated in-use counter. Multiple events can share the same resource. The counter associated with a resource is incremented when a resource becomes associated with a particular event. The counter associated with a resource is decremented when an event completes the use of that particular resource. When the in-use counter for a resource becomes zero, the in-use counter indicates that the resource is unassigned and that the resource can be assigned to a new event.

Method for automatic resource reservation and communication that facilitates using multiple processing events for a single processing task

Circuitry to free the core processor from performing the explicit read operation required to read data into the internal register set. The processor's register set is expanded and a &ldquo;shadow register&rdquo; set is provided. While the core processor is processing one event the &ldquo;context&rdquo; and &ldquo;data&rdquo; and other associated information for the next event is loaded into the shadow register set. When the core processor finishes processing an event, the core processor switches to the shadow register set and it can begin processing the next event immediately. With short service routines, there might not be time to fully pre-fetch the &ldquo;context&rdquo; and &ldquo;data&rdquo; associated with the next event before the current event ends. In this case, the core processor still starts processing the next event and the pre-fetch continues during the event processing. If the core processor accesses a register which is associated with part of the context for which the pre-fetch is still in progress the core processor will automatically stall or delay until the pre-fetch has completed reading the appropriate data.

Enhancing performance by pre-fetching and caching data directly in a communication processor's register set

Circuitry for reading from and writing to memory cells of a group of memory cells. The circuitry comprises read circuitry and write circuitry each connectable to bit lines associated with respective ones of the memory cells. The read circuitry is arranged to read from the cells and the write circuitry is arranged to write to the cells. Wherein the read circuitry and write circuitry are configured so that more cells in the group can be simultaneously written to during a write operation than can be simultaneously read from during a read operation.

Growing community of inventors

Cambridge, MA, United States of America

Duane E Galbi