-
1.发明授权公开(公告)号：US10990561B2
公开(公告)日：2021-04-27
申请号：US15984262
申请日：2018-05-18
Inventor： Shin-Young Ahn , Eun-Ji Lim , Yong-Seok Choi , Young-Choon Woo , Wan Choi
IPC: G06F15/16 , G06F15/173 , G06N3/08 , H04L29/08 , G06F9/50
Abstract: Disclosed herein are a parameter server and a method for sharing distributed deep-learning parameters using the parameter server. The method for sharing distributed deep-learning parameters using the parameter server includes initializing a global weight parameter in response to an initialization request by a master process; performing an update by receiving a learned local gradient parameter from the worker process, which performs deep-learning training after updating a local weight parameter using the global weight parameter; accumulating the gradient parameters in response to a request by the master process; and performing an update by receiving the global weight parameter from the master process that calculates the global weight parameter using the accumulated gradient parameters of the one or more worker processes.
-
公开(公告)号：US10831676B2
公开(公告)日：2020-11-10
申请号：US16165891
申请日：2018-10-19
Inventor： Yong-Seok Choi , Shin-Young Ahn , Eun-Ji Lim , Young-Choon Woo , Wan Choi
IPC: G06F12/1081 , G06F13/28 , G06F13/16
Abstract: Disclosed herein are an apparatus and method for interfacing with common memory. The apparatus for interfacing with common memory includes a computer-input/output (I/O)-interface-protocol-processing unit for receiving a packet for accessing common memory from a computer; a direct memory access unit for transforming the packet into an instruction for performing any one of reading from and writing to the common memory; and a common memory interface unit for transmitting the instruction to the common memory and receiving information about whether execution of the instruction is completed from the common memory.
-
公开(公告)号：US10474643B2
公开(公告)日：2019-11-12
申请号：US15145311
申请日：2016-05-03
Inventor： Myung Hoon Cha , Young Kyun Kim , Hong Yeon Kim , Wan Choi
IPC: G06F16/182 , G06F16/16
Abstract: A distributed file system includes metadata servers and data servers. The metadata server includes a selecting unit to select a data server from the data servers, a chunk allocation requesting unit to request that the selected data server perform chunk allocation, a chunk list managing unit to insert a list of chunks transmitted from the data server into a chunk list pool and determine an arrangement method of the chunk list pool, and a chunk fetching unit to fetch available chunk information from the chunk list pool. The data server includes a receiving unit to receive request for chunk allocation from the metadata server, a chunk allocating unit to allocate chunks in response to the request for chunk allocation and write a list of chunks based on information about the allocated chunk, and a transmitting unit to transmit the list of chunks to the metadata server.
-
公开(公告)号：US10805208B2
公开(公告)日：2020-10-13
申请号：US16215430
申请日：2018-12-10
Inventor： Ock-Kee Baek , Wan Choi
IPC: H04L12/707 , H04L12/24 , H04L12/803 , H04L12/721 , H04L29/06 , H04L29/08
Abstract: Disclosed herein is an apparatus for supporting setting of a network routing path, which includes a sub-cast list management unit for managing a sub-cast list that contains information about one or more other apparatuses for supporting setting of a network routing path with which the apparatus establishes a network session; a session management unit for simultaneously transmitting a SYNC message to at least some of the apparatuses included in the sub-cast list and establishing a network session with one of the apparatuses included in the sub-cast list in consideration of responses to the SYNC message; and a communication unit for transmitting data via the established network session.
-
5.发明授权公开(公告)号：US11487698B2
公开(公告)日：2022-11-01
申请号：US17216322
申请日：2021-03-29
Inventor： Shin-Young Ahn , Eun-Ji Lim , Yong-Seok Choi , Young-Choon Woo , Wan Choi
IPC: G06F15/173 , G06N3/08 , H04L67/10 , G06F9/50
Abstract: Disclosed herein are a parameter server and a method for sharing distributed deep-learning parameters using the parameter server. The method for sharing distributed deep-learning parameters using the parameter server includes initializing a global weight parameter in response to an initialization request by a master process; performing an update by receiving a learned local gradient parameter from the worker process, which performs deep-learning training after updating a local weight parameter using the global weight parameter; accumulating the gradient parameters in response to a request by the master process; and performing an update by receiving the global weight parameter from the master process that calculates the global weight parameter using the accumulated gradient parameters of the one or more worker processes.
-
Patent Agency Ranking
公开(公告)号：US10135926B2
公开(公告)日：2018-11-20
申请号：US14919135
申请日：2015-10-21
Inventor： Ki Sung Jin , Young Chang Kim , Young Kyun Kim , Hong Yeon Kim , Wan Choi
Abstract: Provided herein is a shuffle embedded distributed storage system and method supporting virtual merge, the system and method including a distributed shared storage configured to store a virtual merged file; a plurality of map servers connected to the distributed shared storage via a network, and configured to perform a map function and record a map result data computed as a result of the map function in the distributed shared storage by means of a map result file; and a plurality of reduce servers connected to the distributed shared storage and the map servers via the network, wherein the virtual merged file includes a list of the map result files recorded by the plurality of map servers, and an identifier of a reduce server to which the virtual merged file is to be transmitted.
-
公开(公告)号：US10789083B2
公开(公告)日：2020-09-29
申请号：US15354651
申请日：2016-11-17
Inventor： Dae-Won Kim , Sun-Wook Kim , Jong-Bae Moon , Myeong-Hoon Oh , Byeong-Thaek Oh , Soo-Cheol Oh , Seong-Woon Kim , Ji-Hyeok Choi , Hag-Young Kim , Wan Choi
Abstract: Disclosed herein are a method and apparatus for virtual desktop service. The apparatus includes a connection manager configured to perform an assignment task of assigning a virtual machine to a user terminal using virtual desktop service, a resource pool configured to allocate software resources to a virtual desktop, wherein the software resources include an OS, applications, and user profiles, and a virtual machine infrastructure configured to support hardware resources including a CPU and a memory, wherein the connection manager is configured to perform a coordination task of coordinating a delivery protocol used between the user terminal and servers that provide the virtual desktop service, wherein the resource pool has a management function, wherein the management function is based on usage pattern information about a user's average usage of resources, and wherein the management function uses a physical distance on network from the user terminal to a server.
-
公开(公告)号：US10585726B2
公开(公告)日：2020-03-10
申请号：US15979169
申请日：2018-05-14
Inventor： Eun-Ji Lim , Shin-Young Ahn , Yong-Seok Choi , Young-Choon Woo , Wan Choi
IPC: H04L7/04 , G06F9/54 , G06F9/52 , G06N20/00 , G06F15/167
Abstract: A parameter-sharing apparatus includes a memory allocation unit for managing allocation of a memory area, in which a parameter is to be stored, to a memory box, and updating a mapping table stored in the memory box based on allocation management of the memory area, and an operation processing unit for providing the memory allocation unit with parameter information required for the allocation management of the memory area in which the parameter is to be stored and sharing the parameter stored in the memory box.
-
公开(公告)号：US09804965B2
公开(公告)日：2017-10-31
申请号：US14848841
申请日：2015-09-09
Inventor： Young Chang Kim , Ki Sung Jin , Young Kyun Kim , Hong Yeon Kim , Wan Choi
IPC: G06F12/08 , G06F12/0862 , G06F9/44 , G06F9/455
CPC classification number: G06F12/0862 , G06F9/4401 , G06F9/45558 , G06F2009/45579 , G06F2212/6026
Abstract: A virtual machine host server includes a virtual machine in which a guest operating system is installed and operated, a cache manager for processing at least one of an open request, a close request, and an input/output request for a disk image file of the virtual machine, which is stored in a storage system, and managing a boot workload map and a boot segment, a cache device for caching the boot segment, and a prefetch manager for prefetching the boot segment from the cache device.
-
公开(公告)号：US09760154B2
公开(公告)日：2017-09-12
申请号：US14264214
申请日：2014-04-29
Inventor： Sung Ik Jun , Baik Song An , Jin Ho On , Young Choon Woo , Wan Choi
IPC: G06F1/32
CPC classification number: G06F1/324 , G06F1/3206 , Y02D10/126
Abstract: Provided is a method of dynamically controlling power in a multicore environment including a multicore processor which includes a plurality of cores and a scheduler. The method includes determining whether a management policy is set, collecting frequency change information used to change frequencies of the plurality of cores when it is determined that the management policy is set, calculating an average load of each core on a basis of the frequency change information, calculating an average frequency of each core according to the calculated average load of each core, comparing the average frequency of each core and a predetermined threshold value, and setting a next frequency of each core according to the comparison result.
-
-
-
-
-
-
-
-
-
Search Results
10
records
(took 0.016 seconds)
