GL
G.S. Latsev
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
2 records found
1
The rise of edge computing has enabled a wide range of applications that require decentralized processing and closer interaction with end devices, including telecommunications, industrial automation, distributed multimedia, and IoT systems. In contrast to comparatively homogeneous cloud environments, edge infrastructures often consist of heterogeneous hardware platforms with different architectures, capabilities, and management constraints. This diversity makes experimental evaluation harder to reproduce and operationally more difficult to manage.
This thesis presents the design and implementation of a multi-tenant edge testbed for heterogeneous hardware. The testbed supports both exclusive access to individual devices and shared use of device resources through multiple execution environments. It also supports multiple deployment styles, ranging from direct use of provisioned execution environments to the deployment of additional services on top of them. Target applications include 5G radio access and core network functions, as well as distributed augmented reality pipelines. To support these goals, the system is organized into infrastructure, orchestration, and service layers that address hardware abstraction, multi-tenant experiment management, and user-facing experiment specification and deployment.
The proposed design is evaluated in three ways. First, virtualization choices are assessed on representative devices to determine suitable execution environments for shared and exclusive access. Second, the platform is demonstrated through 5G and distributed augmented reality deployment scenarios. Third, interference under concurrent use is evaluated both on shared hosts and at shared network bottlenecks.
The results indicate that the proposed design provides a practical basis for reproducible multi-tenant experimentation across heterogeneous edge devices and diverse deployment configurations, while also clarifying the remaining limits imposed by shared physical infrastructure under contention. ...
This thesis presents the design and implementation of a multi-tenant edge testbed for heterogeneous hardware. The testbed supports both exclusive access to individual devices and shared use of device resources through multiple execution environments. It also supports multiple deployment styles, ranging from direct use of provisioned execution environments to the deployment of additional services on top of them. Target applications include 5G radio access and core network functions, as well as distributed augmented reality pipelines. To support these goals, the system is organized into infrastructure, orchestration, and service layers that address hardware abstraction, multi-tenant experiment management, and user-facing experiment specification and deployment.
The proposed design is evaluated in three ways. First, virtualization choices are assessed on representative devices to determine suitable execution environments for shared and exclusive access. Second, the platform is demonstrated through 5G and distributed augmented reality deployment scenarios. Third, interference under concurrent use is evaluated both on shared hosts and at shared network bottlenecks.
The results indicate that the proposed design provides a practical basis for reproducible multi-tenant experimentation across heterogeneous edge devices and diverse deployment configurations, while also clarifying the remaining limits imposed by shared physical infrastructure under contention. ...
The rise of edge computing has enabled a wide range of applications that require decentralized processing and closer interaction with end devices, including telecommunications, industrial automation, distributed multimedia, and IoT systems. In contrast to comparatively homogeneous cloud environments, edge infrastructures often consist of heterogeneous hardware platforms with different architectures, capabilities, and management constraints. This diversity makes experimental evaluation harder to reproduce and operationally more difficult to manage.
This thesis presents the design and implementation of a multi-tenant edge testbed for heterogeneous hardware. The testbed supports both exclusive access to individual devices and shared use of device resources through multiple execution environments. It also supports multiple deployment styles, ranging from direct use of provisioned execution environments to the deployment of additional services on top of them. Target applications include 5G radio access and core network functions, as well as distributed augmented reality pipelines. To support these goals, the system is organized into infrastructure, orchestration, and service layers that address hardware abstraction, multi-tenant experiment management, and user-facing experiment specification and deployment.
The proposed design is evaluated in three ways. First, virtualization choices are assessed on representative devices to determine suitable execution environments for shared and exclusive access. Second, the platform is demonstrated through 5G and distributed augmented reality deployment scenarios. Third, interference under concurrent use is evaluated both on shared hosts and at shared network bottlenecks.
The results indicate that the proposed design provides a practical basis for reproducible multi-tenant experimentation across heterogeneous edge devices and diverse deployment configurations, while also clarifying the remaining limits imposed by shared physical infrastructure under contention.
This thesis presents the design and implementation of a multi-tenant edge testbed for heterogeneous hardware. The testbed supports both exclusive access to individual devices and shared use of device resources through multiple execution environments. It also supports multiple deployment styles, ranging from direct use of provisioned execution environments to the deployment of additional services on top of them. Target applications include 5G radio access and core network functions, as well as distributed augmented reality pipelines. To support these goals, the system is organized into infrastructure, orchestration, and service layers that address hardware abstraction, multi-tenant experiment management, and user-facing experiment specification and deployment.
The proposed design is evaluated in three ways. First, virtualization choices are assessed on representative devices to determine suitable execution environments for shared and exclusive access. Second, the platform is demonstrated through 5G and distributed augmented reality deployment scenarios. Third, interference under concurrent use is evaluated both on shared hosts and at shared network bottlenecks.
The results indicate that the proposed design provides a practical basis for reproducible multi-tenant experimentation across heterogeneous edge devices and diverse deployment configurations, while also clarifying the remaining limits imposed by shared physical infrastructure under contention.
Program synthesis remains largely unexplored in the context of playing games, where exploration and exploitation are crucial for solving tasks within complex environments. FrAngel is a program synthesis algorithm that addresses both of these aspects with its fragments used for the generation of new candidate programs. We introduce a generalised version of the FrAngel program synthesis algorithm that accepts an arbitrary iterator and grammar, enabling flexible modifications. We then formulate Program by Example (PBE) program synthesis from rewards and apply this framework to Minecraft's navigation task, where dense rewards guide the algorithm's exploration. Then, we use the generalised version of the algorithm to conduct experiments in the context of exploration of the algorithm. The experiments show the importance of exploration as a step required for the exploitation and finding the wanted solution to the task while also revealing that sometimes more exploration does not necessarily mean reaching the solution faster.
...
Program synthesis remains largely unexplored in the context of playing games, where exploration and exploitation are crucial for solving tasks within complex environments. FrAngel is a program synthesis algorithm that addresses both of these aspects with its fragments used for the generation of new candidate programs. We introduce a generalised version of the FrAngel program synthesis algorithm that accepts an arbitrary iterator and grammar, enabling flexible modifications. We then formulate Program by Example (PBE) program synthesis from rewards and apply this framework to Minecraft's navigation task, where dense rewards guide the algorithm's exploration. Then, we use the generalised version of the algorithm to conduct experiments in the context of exploration of the algorithm. The experiments show the importance of exploration as a step required for the exploitation and finding the wanted solution to the task while also revealing that sometimes more exploration does not necessarily mean reaching the solution faster.