Intel Corporation

Alejandro Duran Gonzalez

Francesc Guim Bernat

Kshitij Arun Doshi

Karthik Kumar

Thomas Willhalm

Raj K Ramanujan

Daniel Rivas Barragan

Harald Servat

Arch D Robison

Technologies for dynamically sharing remote resources include a computing node that sends a resource request for remote resources to a remote computing node in response to a determination that additional resources are required by the computing node. The computing node configures a mapping of a local address space of the computing node to the remote resources of the remote computing node in response to sending the resource request. In response to generating an access to the local address, the computing node identifies the remote computing node based on the local address with the mapping of the local address space to the remote resources of the remote computing node and performs a resource access operation with the remote computing node over a network fabric. The remote computing node may be identified with system address decoders of a caching agent and a host fabric interface. Other embodiments are described and claimed.

Technologies for dynamically sharing remote resources across remote computing nodes

Method and apparatus implementing smart store operations with conditional ownership requests. One aspect includes a method implemented in a multi-core processor, the method comprises: receiving a conditional read for ownership (CondRFO) from a requester in response to an execution of an instruction to modify a target cache line (CL) with a new value, the CondRFO identifying the target CL and the new value; determining from a local cache a local CL corresponding to the target CL; determining a local value from the local CL; comparing the local value with the new value; setting a coherency state of the local CL to (S)hared when the local value is same as the new value; setting the coherency state of the local CL to (I)nvalid when the local value is different than the new value; and sending a response and a copy of the local CL to the requester. Other embodiments include an apparatus configured to perform the actions of the methods.

Method and apparatus for smart store operations with conditional ownership requests

Examples may include techniques for collective operations in a distributed architecture. A collective operation request message from a computing node causes collective operations at one or more target computing nodes communicatively coupled with the computing node through a network switch. The collective operation request message also causes the network switch to perform collective operations on collective operation results received from the one or more target computing nodes.

Method and apparatus for compute end point based collective operations

Technologies for performing switch-based collective operations in a fabric architecture include a network switch communicatively coupled to a plurality of computing nodes. The network switch is configured to identify sub-operations of a collective operation of a collective operation request received from one of the computing nodes and identify a plurality of operands for each of the sub-operations. The network switch is additionally configured to request a value for each of the operands from a corresponding target computing node at which the respective value is stored, determine a result of the collective operation as a function of the requested operand values, and transmit the result to the requesting computing node. Other embodiments are described herein.

Technologies for performing switch-based collective operations in distributed architectures

Method and apparatus implementing smart store operations with conditional ownership requests. One aspect includes a method implemented in a multi-core processor, the method comprises: receiving a conditional read for ownership (CondRFO) from a requester in response to an execution of an instruction to modify a target cache line (CL) with a new value, the CondRFO identifying the target CL and the new value; determining from a local cache a local CL corresponding to the target CL; determining a local value from the local CL; comparing the local value with the new value; setting a coherency state of the local CL to (S)hared when the local value is same as the new value; setting the coherency state of the local CL to (I)nvalid when the local value is different than the new value; and sending a response and a copy of the local CL to the requester. Other embodiments include an) apparatus configured to perform the actions of the methods.

Methods and apparatuses relating to translating a logical thread identification to a physical thread identification. A processor may include a plurality of cores that include a buffer, and a thread mapping hardware unit to: return a physical thread identification in response to a logical thread identification sent to a buffer of a first core when the buffer includes a logical to physical thread mapping for the logical thread identification, and send a request to the buffers of the other cores when the first core's buffer does not include the logical to physical thread mapping for the logical thread identification, wherein each of the other cores are to send an unknown identification response if their buffer does not include the logical thread identification and at least one of the other cores is to send the physical thread identification to the first core if its buffer includes the logical thread identification.

Apparatuses and methods to translate a logical thread identification to a physical thread identification

Systems and methods may provide for identifying a nested loop iteration space in user code, wherein the nested loop iteration space includes a plurality of outer loop iterations, and distributing iterations from the nested loop iteration space across a plurality of threads, wherein each thread is assigned a group of outer loop iterations. Additionally, a compiler output may be automatically generated, wherein the compiler output contains serial code corresponding to each group of outer loop iterations and de-linearization code to be executed outside the plurality of outer loop iterations. In one example, the de-linearization code includes index recovery code that is positioned before one or more instances of the serial code in the compiler output.

Loop nest parallelization without loop linearization

Data and a memory address associated with the data may be received. A hash value of the memory address may be calculated by using a first hash function. The data may be stored at a cache set of a plurality of cache sets of a cache memory based on the hash value calculated from the first hash function. A determination may be made as to whether the storing of the data at the cache set of the plurality of cache sets of the cache memory is associated with a conflict ratio of the cache memory exceeding a threshold ratio. In response to the conflict ratio exceeding the threshold ratio, a second hash value of a second memory address associated with a second data may be calculated by using a second hash function that is different than the first hash function.

Changing a hash function based on a conflict ratio associated with cache sets

Technologies for multithreaded synchronization and work stealing include a computing device executing two or more threads in a thread team. A thread executes all of the tasks in its task queue and then exchanges its associated task stolen flag value with false and stores that value in a temporary flag. Subsequently, the thread enters a basic synchronization barrier. The computing device performs a logical-OR reduction over the temporary flags of the thread team to produce a reduction value. While waiting for other threads of the thread team to enter the barrier, the thread may steal a task from a victim thread and set the task stolen flag of the victim thread to true. After exiting the basic synchronization barrier, if the reduction value is true, the thread repeats exchanging the task stolen flag value and entering the basic synchronization barrier. Other embodiments are described and claimed.

Growing community of inventors

Barcelona, Spain

Alejandro Duran Gonzalez