-
1.发明公开公开(公告)号：US20240118878A1
公开(公告)日：2024-04-11
申请号：US18542544
申请日：2023-12-15
发明人： Jaejin Lee , Jungho Park , Gangwon Jo , Heehoon Kim , Jinpyo Kim
摘要： A method for determining optimization applicability on an intermediate representation from a program is performed by one or more processors, and includes receiving, as a query, a subgraph of the intermediate representation that is a subject of determination of optimization applicability, determining a validity of the query, and if the query is valid, determining optimization applicability on the subgraph, in which the program includes data and a plurality of operations, and the intermediate representation includes a plurality of data nodes, a plurality of operation nodes, and a plurality of edges representing input/output relationships between the plurality of data nodes and the plurality of operation nodes.
-
公开(公告)号：US20240118876A1
公开(公告)日：2024-04-11
申请号：US18542563
申请日：2023-12-15
发明人： Jaejin Lee , Jungho Park , Gangwon Jo , Heehoon KIM , Jinpyo KIM
IPC分类号： G06F8/41
CPC分类号： G06F8/41
摘要： A method for managing an intermediate representation from a program is executed by one or more processors, and includes extracting, from the program, information on data for input and output and information on operation, generating an intermediate representation from the program using the extracted information on data and the extracted information on operation, storing, in a database, a corresponding relationship between the program and the intermediate representation, storing execution information on operation of the intermediate representation, and deleting at least a part of the intermediate representation based on the execution information.
-
公开(公告)号：US11720405B2
公开(公告)日：2023-08-08
申请号：US17145958
申请日：2021-01-11
发明人： Wookeun Jung , Jaejin Lee , Seung Wook Lee
CPC分类号： G06F9/5027 , G06F9/30014 , G06F9/3802 , G06F9/3836 , G06F9/48 , G06N3/063
摘要： A processor-implemented accelerator method includes: reading, from a memory, an instruction to be executed in an accelerator; reading, from the memory, input data based on the instruction; and performing, on the input data and a parameter value included in the instruction, an inference task corresponding to the instruction.
-
4.发明授权公开(公告)号：US10725667B2
公开(公告)日：2020-07-28
申请号：US15874322
申请日：2018-01-18
发明人： Jaejin Lee , Gangwon Jo
IPC分类号： G06F3/06 , G06F15/163 , G06F15/167 , G06F12/08 , G06F12/14
摘要： Disclosed herein are a method of transferring data in a parallel system including a main device and at least one accelerator, and a parallel system for performing the method. The method of transferring data in a heterogeneous system including a main device and at least one accelerator includes: turning off a write permission for a first main memory area corresponding to a first accelerator memory area where input data for a computation task is stored; performing the computation task by using the at least one accelerator; and turning off a read permission for a second main memory area corresponding to a second accelerator memory area where output data for the computation task is stored, in the state in which data of the second accelerator memory area has not been transferred to the second main memory area.
-
公开(公告)号：US09367291B2
公开(公告)日：2016-06-14
申请号：US14228868
申请日：2014-03-28
发明人： Jin-Seok Lee , Seong-Gun Kim , Dong-Hoon Yoo , Seok-Joong Hwang , Jeongho Nah , Jaejin Lee , Jun Lee
IPC分类号： G06F9/45
CPC分类号： G06F8/40
摘要： An apparatus and method for generating vector code are provided. The apparatus and method generate vector code using scalar-type kernel code, without user's changing a code type or modifying data layout, thereby enhancing user's convenience of use and retaining the portability of OpenCL.
摘要翻译： 提供了一种用于产生矢量码的装置和方法。 该设备和方法使用标量型内核代码生成向量代码，无需用户更改代码类型或修改数据布局，从而提高用户使用方便性并保留OpenCL的可移植性。
-
6.发明授权Cluster system based on parallel computing framework, and host node, computing node and method for executing application therein 有权基于并行计算框架的集群系统，主机节点，计算节点及其中执行应用的方法公开(公告)号：US09485303B2
公开(公告)日：2016-11-01
申请号：US13734407
申请日：2013-01-04
发明人： Jaejin Lee , Jungwon Kim
IPC分类号： G06F15/177 , H04L29/08 , G06F9/54
摘要： A cluster system based on a parallel computing framework is provided, and the cluster system includes a host node configured to execute a host program for a parallel computing framework and a computing node configured to be connected to the host node and execute a kernel program for the parallel computing frame work.
摘要翻译： 提供了一种基于并行计算框架的集群系统，并且集群系统包括被配置为执行用于并行计算框架的主机程序的主机节点和被配置为连接到主机节点的计算节点，并执行用于 并行计算框架工作。
-
7.发明授权公开(公告)号：US09208088B2
公开(公告)日：2015-12-08
申请号：US13733396
申请日：2013-01-03
发明人： Jaejin Lee , Junghyun Kim
CPC分类号： G06F12/0815 , G06F12/0837 , G06F12/109 , G06F2212/656
摘要： A shared virtual memory management apparatus for ensuring cache coherence. When two or more cores request write permission to the same virtual memory page, the shared virtual memory management apparatus allocates a physical memory page for the cores to change data in the allocated physical memory page. Thereafter, changed data is updated in an original physical memory page, and accordingly it is feasible to achieve data coherence in a multi-core hardware environment that does not provide cache coherence.
摘要翻译： 一种用于确保高速缓存一致性的共享虚拟内存管理装置。 当两个或多个核心向相同的虚拟内存页面请求写入权限时，共享虚拟内存管理装置为核心分配物理存储器页面以改变所分配的物理存储器页面中的数据。 此后，在原始物理存储器页面中更新改变的数据，因此在不提供高速缓存一致性的多核硬件环境中实现数据一致性是可行的。
-
公开(公告)号：US12106076B2
公开(公告)日：2024-10-01
申请号：US17853090
申请日：2022-06-29
申请人： MOREH CORP.
发明人： Jaejin Lee , Wookeun Jung
摘要： The present disclosure relates to a method for generating a program for use in an accelerator for deep learning. The method may include receiving, by a computing device, a deep learning application, generating an element-wise operation list included in the deep learning application, generating an intermediate expression from the element-wise operation list, and generating, based on the intermediate expression, a program for use in an accelerator for the deep learning application.
-
公开(公告)号：US20220326922A1
公开(公告)日：2022-10-13
申请号：US17853028
申请日：2022-06-29
申请人： MOREH CORP.
发明人： Jaejin Lee , Wookeun Jung
摘要： The present disclosure relates to a method for automatically optimizing a program based on reinforcement learning. The method for automatically optimizing a program based on reinforcement learning includes (a) receiving an input for a source program, which includes a fixed parameter and variable parameter, (b) generating the source program based on the received input, (c) converting the source program into an object program, (d) executing the converted object program to measure a performance of the executed object program, (e) inputting the variable parameter and the measured performance into a machine learning model, and outputting a variation of the variable parameter, and (f) regenerating a source program reflecting the variation of the variable parameter.
-
公开(公告)号：US20230031226A1
公开(公告)日：2023-02-02
申请号：US17964626
申请日：2022-10-12
申请人： MOREH CORP.
发明人： Jaejin Lee , Jungho Park , Youngdong Do
IPC分类号： G06N3/08
摘要： Provided is a method for processing a deep learning task through a deep learning framework. The method may include executing, by a computing device, a deep learning task on a deep learning framework, determining at least one of a primary accelerator or a secondary accelerator to execute the deep learning task, allocating the deep learning task to at least one of the determined primary accelerator or secondary accelerator, and generating, based on a result processed by at least one of the determined primary accelerator or secondary accelerator, result data for the deep learning task. The secondary accelerator may be an accelerator heterogeneous to the primary accelerator.
-
-
-
-
-
-
-
-
-
搜索结果
12 条记录 (耗时 0.021 秒)
