公开(公告)号：EP4398102A1

公开(公告)日：2024-07-10

申请号：EP23217957.2

申请日：2023-12-19

申请人： Tata Consultancy Services Limited

发明人： PHALAK, Chetan Dnyandeo ,  CHAHAL, Dheeraj ,  SINGHAL, Rekha

IPC分类号： G06F9/50

CPC分类号： G06F9/505 ,  G06F9/5083 ,  G06F9/5072 ,  G06F2209/50820130101

摘要： Cloud and Fog computing are complementary technologies used for complex Internet of Things (IoT) based deployment of applications. With an increase in the number of internet-connected devices, the volume of data generated and processed at higher speeds has increased substantially. Serving a large amount of data and workloads for predictive decisions in real-time using fog computing without Service-Level Objective (SLO) violation is a challenge. Present disclosure provides systems and method for inference management wherein a suitable execution workflow is automatically generated to execute machine learning (ML)/deep learning (DL) inference requests using fog with various type of instances (e.g., Function-as-a-Service (FaaS) instance, Machine Learning-as-a-service (MLaaS) instance, and the like) provided by cloud vendors/platforms. Generated workflow minimizes the cost of deployment as well as SLO violations.

公开(公告)号：EP3803596B1

公开(公告)日：2024-06-05

申请号：EP19840262.0

申请日：2019-07-23

IPC分类号： G06F9/50 ,  G06F15/16 ,  G06F1/3206 ,  G06F1/329

CPC分类号： G06F1/3206 ,  G06F1/329 ,  G06F9/5083 ,  G06F2209/501120130101 ,  G06F9/5094

公开(公告)号：EP3408742A1

公开(公告)日：2018-12-05

申请号：EP16702504.8

申请日：2016-01-26

申请人： Telefonaktiebolaget LM Ericsson (PUBL)

发明人： GÉHBERGER, Dániel ,  MÁTRAY, Péter ,  NÉMETH, Gábor

IPC分类号： G06F9/50 ,  G06F9/54 ,  H04L29/08

CPC分类号： G06F9/546 ,  G06F9/5083 ,  G06F2209/548 ,  H04L67/1008

摘要： A technique for determining of a load of an application in a cloud computing environment (100) is disclosed. The application is executed with one or more application instances (102) in the cloud computing environment (100), wherein each of the one or more application instances (102) obtains input data from a respective input queue (104). A method implementation for supporting the technique comprises determining a wait indicator for at least one of the one or more application instances (102), the wait indicator for an application instance (102) indicating a relation between empty states of the input queue (104) of the application instance (102) and non-empty states of the input queue (104) of the application instance (102), and triggering forwarding of the wait indicator determined for the at least one of the one or more application instances (102) to a load determination component (106).

公开(公告)号：EP3392765A1

公开(公告)日：2018-10-24

申请号：EP18159607.3

申请日：2018-03-01

申请人： INTEL Corporation

发明人： VEMBU, Balaji ,  MASTRONARDE, Josh B. ,  KABURLASOS, Nikos

IPC分类号： G06F9/50

CPC分类号： G06F1/3287 ,  G06F1/3296 ,  G06F9/4881 ,  G06F9/5083 ,  G06F9/5088 ,  G06F13/4022

摘要： In an example, an apparatus comprises logic, at least partially comprising hardware logic, to power on a first set of processing clusters, dispatch a workload to the first set of processing clusters, detect a full operating state of the first set of processing clusters, and in response to the detection of a full operating state of the first set of processing clusters, to power on a second set of processing clusters. Other embodiments are also disclosed and claimed.

公开(公告)号：EP2953399B1

公开(公告)日：2018-10-24

申请号：EP14290160.2

申请日：2014-06-02

申请人： Alcatel Lucent

发明人： Haldrun, Grob-Lipski ,  Bernd, Haberland

IPC分类号： H04W28/08 ,  H04W88/08 ,  H04W28/02

CPC分类号： H04L47/24 ,  G06F9/5038 ,  G06F9/505 ,  G06F9/5083 ,  H04L43/0882 ,  H04L43/16 ,  H04L47/125 ,  H04L49/50 ,  H04L67/1008 ,  H04L67/1012 ,  H04W24/08 ,  H04W28/08 ,  H04W28/18 ,  H04W72/0486 ,  H04W72/0493 ,  H04W72/085 ,  H04W72/1252 ,  H04W72/1257 ,  H04W88/085

摘要： It is proposed a method for operating a radio communication network or a network element thereof, comprising the steps of: determining (2) a bearer to be processed in the radio communication network; selecting (4) a baseband unit, BBU, out of a plurality of BBUs, which satisfies at least one property of the bearer; and placing (8) the bearer on the selected BBU for processing.

公开(公告)号：EP3384389A1

公开(公告)日：2018-10-10

申请号：EP16810204.4

申请日：2016-11-25

申请人： Microsoft Technology Licensing, LLC

发明人： PURI, Ranjit ,  SCHULTZ, Benjamin M. ,  THALER, David G.

IPC分类号： G06F9/50

CPC分类号： H04L67/1029 ,  G06F9/5083 ,  H04L43/0805 ,  H04L67/2842

摘要： A load balancer receives a client request from a client device for a connection to an application. The load balancer queries a monitoring server for a list of one or more application servers associated with the application. The monitoring server determines, based on cache state information of the plurality of application servers, the list of one or more application servers. The load balancer establishes a connection on behalf of the client device to one of the application servers.

公开(公告)号：EP3374889A1

公开(公告)日：2018-09-19

申请号：EP17729607.6

申请日：2017-05-24

申请人： Google Inc.

发明人： MEHTA, Aranyak ,  TAN, Xi ,  HARRIS, Christopher ,  IEONG, Samuel ,  CONSTANTIN, Gabriel

IPC分类号： G06F17/30 ,  H04L29/08 ,  G06Q30/02

CPC分类号： H04L47/125 ,  G06F9/5083 ,  G06Q30/02 ,  G06Q30/0241 ,  H04L67/02

摘要： Systems and methods of a balancing data requests over a computer network are provided. The system can receive a request for content to be rendered in a content slot on a computing device. The content slot can be coded with a default timer value. The system can execute a real-time content selection process to identify content items that each have a magnitude. The system can determine a timer value based on the magnitudes, and transmit the timer value to the computing device to override the default timer value to cause the computing device to replace a first content item rendered in the content slot upon expiration of the timer.

公开(公告)号：EP3373556A1

公开(公告)日：2018-09-12

申请号：EP18158966.4

申请日：2018-02-27

申请人： THOMSON Licensing

发明人： LEGALLAIS, Yvon ,  ONNO, Stephane ,  LE SCOUARNEC, Nicolas

IPC分类号： H04L29/08 ,  G06F9/50 ,  G06F9/46 ,  H04L12/24

CPC分类号： H04L41/0823 ,  G06F9/45558 ,  G06F9/466 ,  G06F9/5077 ,  G06F9/5083 ,  G06F9/5088 ,  G06F2009/45595 ,  H04L41/0896 ,  H04L41/5054 ,  H04L41/5096 ,  H04L47/125 ,  H04L67/10 ,  H04L67/1031

摘要： Network element (120) adapted to belong to a network equipment (100) configured to operate a plurality of network functions (Fx, Fy, Fz) and to be in communication with one or more devices (10), wherein the network element (120) comprises:
- one or more interfaces of communication (402) configured to:
▪ receive a first packet after processing by a first instance of a network function (Fx), said first packet comprising an indicator of the oldest pending transaction handled by the first instance of said network function (Fx);
▪ receive a second packet after processing by a second instance of said network function (Fx), said second packet comprising an indicator of the oldest pending transaction handled by the second instance of said network function (Fx), the first packet and the second packet comprising the same information before respectively processing by the first and second instances of said network function (Fx);

- a comparator (122) to determine whether the indicator of the first packet is equal to the indicator of the second packet,
and wherein said one or more interfaces of communication (402) are further configured to notify a controller (101) that the first and second indicators are equal in order to discard the former operated instance of said network function (Fx) amongst the first and the second instances.

公开(公告)号：EP3371701A1

公开(公告)日：2018-09-12

申请号：EP16795477.5

申请日：2016-10-28

申请人： Chicago Mercantile Exchange, Inc.

发明人： HELLEREN, Erik

IPC分类号： G06F11/07 ,  G06Q40/04 ,  G06F9/48 ,  G06F9/50

CPC分类号： G06F11/1441 ,  G06F9/4856 ,  G06F9/505 ,  G06F9/5072 ,  G06F9/5083 ,  G06F9/5088 ,  G06F11/1474 ,  G06F2201/805 ,  G06F2201/85

摘要： A computer implemented method for providing fault tolerance to a plurality of machines includes determining an aggregate load for each surviving machine of a plurality of surviving machines; calculating a recovery load of one or more orphaned jobs resulting from a terminating event; and selecting to recover and perform one or more of the orphaned jobs, by one or more of the surviving machines, based upon (i) the recovery load of the one or more orphaned jobs; (ii) the job load of the one or more orphaned jobs; and (iii) the aggregate loads of the surviving machines.

公开(公告)号：EP2583417B1

公开(公告)日：2018-08-29

申请号：EP10853042.9

申请日：2010-06-18

申请人： Nokia Solutions and Networks Oy

发明人： FU, Bin ,  MAO, Fengyu ,  WANG, Yongming ,  ZHANG, Ruoyuan

IPC分类号： H04L29/08 ,  H04L29/14 ,  G06F9/50

CPC分类号： H04L41/00 ,  G06F9/5083 ,  G06F11/1484 ,  G06F11/202 ,  H04L45/22 ,  H04L45/28 ,  H04L67/1023 ,  H04L67/1029 ,  H04L67/1034 ,  H04L67/28 ,  H04L69/40

摘要： A server cluster is described, which enables load balancing between servers in the cluster. At least some of the servers in the cluster are divided into a plurality of virtual servers, wherein each virtual server is associated with a neighbouring server, which neighbouring server acts as a backup for that virtual server. The neighbouring server of each virtual server of a particular server is part of a different physical server to the virtual server, such that in the event that a physical server is unavailable for use, the load of the virtual servers of that physical server is split between a number of different physical servers, thereby reducing the likelihood of overloading any particular physical server.