Future Wei Technologies, Inc.

Samsung Electronics Co., Ltd.

Other

Huawei Technologies Co., Limited

University of Chicago

Chen Tian

Ziang Hu

Daniel Waddington

Reza Yazdani

Zongfang Lin

Tongping Liu

Liya Chen

Feng Ye

Handong Ye

Shen Chi Chen

Haryadi Gunawi

Shan Lu

Haopeng Liu

Guangpu Li

Lawrence Davison

Xu Wang

Jilong Kuang

Robert Lasater

Yan Zeng

Wei Ling

Jun Wang

Ziang Zu

Keval Vora

Jeffrey Lukman

Viet Trinh

Huan Ke

Karthik Rao

Sharanyan Srikanthan

A method of distributed resource management in a distributed computing system includes determining usage of respective hardware resources by an application and generating usage metrics for the application, and assigning the application to a cluster of hardware resources to optimize diversity of usage of hardware resources in the cluster and to enhance utilization of the hardware resources by applications running in that cluster. The diversity of usage of the hardware resources is determined from respective usage metrics of the respective applications running in that cluster. The diversity of usage of the hardware resources in the cluster is optimized by assigning the application to a diversity pool of hardware resources adapted to minimize interference when applications assigned to the diversity pool of hardware resources access the hardware resources in the diversity pool and assigning applications from different diversity pools to the cluster of hardware resources.

Distributed resource management by improving cluster diversity

A method of controlling an electronic device is provided, comprising receiving a selection contact within a first icon region of a first icon displayed on a pressure-sensing display screen of the electronic device, displaying on the pressure-sensing display screen a second set of icons related to the first icon, detecting that the selection contact uninterruptedly moves to a second icon of the second set of icons and selects the second icon, displaying on the pressure-sensing display screen a third set of icons related to the second icon, detecting that the selection contact uninterruptedly moves to a third icon of the third set of icons and selects the third icon, and initiating an application or function associated with the third icon.

Software defined icon interactions with multiple and expandable layers

The disclosure relates to technology for establishing communication between a remote device and a virtual machine on a computing node of the cloud-based services via the distributed network. The cloud-based services include multiple operating systems and one or more applications. In response to receiving an instruction from the remote device the virtual machine is accessed, where the virtual machine is associated with at least one of a user of the remote device and the remote device. When directed by an instruction from the remote device, the virtual machine identifies one of the operating systems in which to execute the one or more applications and transmits results, based on execution of the one or more applications by the virtual machine, to the remote device via the distributed network.

Cloud based phone services accessible in the cloud by a remote device

A sender device includes a non-transitory memory storage comprising instructions and a location control policy, and a processor coupled to the memory. The processor executes the instructions to generate an email, generate a control mechanism for the email, wherein the control mechanism instructs a security server to implement the location control policy and wherein the location control policy affects a recipient device's use of the email, and integrate the control mechanism into the email to generate an integrated email. The sender device further includes a transmitter coupled to the processor and configured to transmit the integrated email to the security server for the security server to implement the control mechanism.

Location control and access control of emails

A method implemented by a network element (NE) in a distributed system, the method comprising tracing an execution of a program in the distributed system to produce a record of the execution of the program, wherein the record indicates states of shared resources at various times during the execution of the program, identifying a vulnerable operation that occurred during the program execution based on the record, wherein the record indicates that a first shared resource of the shared resources is in a flawed state after a node that caused the first shared resource to be in the flawed state crashed, and determining that the vulnerable operation results in a time of fault (TOF) bug based on performing a fault-tolerance mechanism.

Automatically detecting time-of-fault bugs in cloud systems

A device comprises a non-transitory memory storage comprising instructions, a network interface, and one or more processors in communication with the memory storage and the network interface. The one or more processors execute the instructions to receive, via the network interface, a write request from a client device; send, via the network interface, the received write request to a set of follower devices; receive, via the network interface, a first acknowledgment from a majority of the follower devices of the set of follower devices that the write request was processed; send, via the network interface, the received write request to a set of peer devices; receive, via the network interface, second acknowledgments from a majority of the set of peer devices that the write request was processed; and send an acknowledgment of the write request to the client device in response to the receipt of the first acknowledgments and the second acknowledgments.

Cross-data center hierarchical consensus scheme with geo-aware leader election

A method implemented by a cloud computing devices comprises removing, by the cloud computing device, data associated with a function from an execution data structure in response to determining that the function is waiting for an input event, adding, by the cloud computing device, a context associated with the function to a management data structure while the function is waiting for the input event, the context associated with the function comprising software components associated with the function and an intermediate variable associated with the function, executing, by the cloud computing device, the function with the input event in response to receiving the input event, and removing, by the cloud computing device, the context associated with the function from the management data structure in response to receiving the input event.

Auto-adaptive serverless function management

A first device comprises: a memory configured to store a first sub-graph that is part of a distributed graph associated with a distributed graph processing network; a processor coupled to the memory and configured to: process the first sub-graph; and save, independently of a second device in the distributed graph processing network, a first snapshot of a first execution state of the first device at a first iteration time; and a transmitter coupled to the processor and configured to transmit the first snapshot to the second device or to a third device.

Fault tolerance in distributed graph processing networks

A method for detecting distributed concurrency errors in a distributed cloud computing system includes tracing operations that access objects in functions involving inter-process messaging, applying a set of happens-before rules to the traced operations. Analyzing the traced operations to identify concurrent operations that access a common object to generate a list of potential distributed concurrency errors (DCbugs). Pruning the list of DCbugs to remove DCbugs having only local effect and that do not generate run-time errors.

Automatically detecting distributed concurrency errors in cloud systems

Described herein are systems and methods for distributed concurrency (DC) bug detection. The method includes identifying a plurality of nodes in a distributed computing cluster; identifying a plurality of messages to be transmitted during execution of an application by the distributed computing cluster; determining a set of orderings of the plurality of messages for DC bug detection, the set of orderings determined based upon the plurality of nodes and the plurality of messages; removing a subset of the orderings from the set of orderings based upon one or more of a state symmetry algorithm, a disjoint-update independence algorithm, or a zero-crash-impact reordering algorithm; and performing DC bug detection testing using the set of orderings after the subset of the orderings is removed from the set of orderings.

Model checker for finding distributed concurrency bugs

System and method embodiments are provided for creating data structure for parallel programming. A method for creating data structures for parallel programming includes forming, by one or more processors, one or more data structures, each data structure comprising one or more global containers and a plurality of local containers. Each of the global containers is accessible by all of a plurality of threads in a multi-thread parallel processing environment. Each of the plurality of local containers is accessible only by a corresponding one of the plurality of threads. A global container is split into a second plurality of local containers when items are going to be processed in parallel and two or more local containers are merged into a single global container when a parallel process reaches a synchronization point.

Systems and methods for creating and using a data structure for parallel programming

A device for tracing a computer program comprises a non-transitory memory having instructions and one or more processors in communication with the memory. The one or more processors execute the instructions to generate trace code from a trace specification program to trace the computer program. The one or more processors calculate values by executing the computer program and the trace code records, in parallel, the values at a predetermined location in execution of the computer program.

Parallel tracing

In one embodiment, a method of false sharing detection includes performing, by a device, a plurality of optimization passes on source code, to produce optimized source code and receiving, by the device, selection criteria, The method also includes adding instrumentation to the optimized source code, by the device, after performing the plurality of optimization passes, to produce an instrumented code, where the instrumentation is configured to track memory access addresses and access types of global variables and heap variables in accordance with the selection criteria.

System and method for detecting false sharing

In one embodiment, a method for predicting false sharing includes running code on a plurality of cores and determining whether there is potential false sharing between a first cache line and a second cache line, and where the first cache line is adjacent to the second cache line. The method also includes tracking the potential false sharing and reporting the potential false sharing.

System and method for false sharing prediction

A method for operating a multithread processing system is provided, including assigning, by a controller, a subset of a plurality of tasks to a plurality of threads during a time N, collecting, by the controller, data during the time N concerning the operation of the plurality of threads, analyzing, by the controller, the data to determine at least one condition concerning the operation of the plurality of threads during the time N, and adjusting, by the controller, a number of the plurality of threads available in time N+1 in accordance with the at least one condition.

Growing community of inventors

Union City, CA, United States of America

Chen Tian