distributedsystems play an increasingly important role in leading-edge networks with high availability requirements, including software-defined networks (SDN), where replicating essential network state information is...
详细信息
ISBN:
(纸本)9781538694435
distributedsystems play an increasingly important role in leading-edge networks with high availability requirements, including software-defined networks (SDN), where replicating essential network state information is critical to ensure resilience under failures. distributed consensus based strong consistency algorithms, such as Raft, are often used to ensure that all components of the distributed system agree on their view of the replicated data, even when a minority of the distributed components crash. Another critical requirement for highly available networks is to gracefully handle overload conditions, where the demands on the network exceed expected levels for a period of time, such as during natural or man-made disasters or popular sporting events. Hence, the strong consistency algorithms used in such networks must also behave gracefully under overload conditions. We show that, in fact, strong consistency algorithms such as Raft may not behave gracefully under overload conditions and can in fact significantly negatively affect SDN control plane availability in these circumstances. We demonstrate that the open-source ONOS SDN controller, which uses the Java-based Atomix implementation of Raft, exhibits such behavior under intent overload, resulting in the loss of requests to the network, and with the entire SDN network eventually crashing. We further demonstrate similar behaviors of the Python-based pysyncobj implementation of Raft. We then propose DynRaft, a dynamic add-on to Raft implementations that continues to ensure the formally proven strong consistency properties of Raft, and demonstrate the effectiveness of DynRaft with the pysyncobj implementation under emulated overload conditions.
Advances in field reconfigurable technology have made possible the design and implementation of highly flexible parallel multi-processor-memory systems;system reliability is often an important measure of these systems...
详细信息
Legacy systems are business-critical systems that have been developed over the past decades. These systems are inflexible and have many issues. Therefore, they need to be modernised and migrated into new technological...
详细信息
ISBN:
(纸本)9781538621622
Legacy systems are business-critical systems that have been developed over the past decades. These systems are inflexible and have many issues. Therefore, they need to be modernised and migrated into new technological environments to reduce costs and to improve flexibility. Thus, a systematic review is conducted in this paper to consider legacy system modernisation strategies, which range from simple wrapping to full migration. The quality aspects of the migration process need to be considered when selecting a moderation strategy to satisfy the migration requirements. Therefore, this paper proposes the integration of quality into migration processes, which has a considerable impact on cost and risk.
The author has been teaching parallel computing since the late 1990s. This paper provides a brief history of some of the key experiences in his journey as a parallel computing educator, including formative faculty dev...
详细信息
ISBN:
(纸本)9781728174457
The author has been teaching parallel computing since the late 1990s. This paper provides a brief history of some of the key experiences in his journey as a parallel computing educator, including formative faculty development workshops, technology shifts, NSF-sponsored projects, and courses taught. By reading this history, the reader may see how much has changed over the past two decades, and so gain insight into how to prepare for changes to parallel computing education in the future.
We have previously proposed PTIDES (Programming Temporally Integrated distributed Embedded systems), a discrete-event framework that binds real-time with model lime at sensors, actuators, and network interfaces. In th...
详细信息
ISBN:
(纸本)9780769534251
We have previously proposed PTIDES (Programming Temporally Integrated distributed Embedded systems), a discrete-event framework that binds real-time with model lime at sensors, actuators, and network interfaces. In this experimental effort we focus on performance issues and tradeoffs in PTIDES implementation. We address event processing performance with respect to other distributed discrete-event approaches that can be applied in a similar setting. The procedure is experimentally, evaluated on a distributed setup with standard software and networking components.
When real-time embedded systems are constructed using conventional software architectures, it is often practically impossible to verify their timing properties. This paper describes a software architecture based on pr...
详细信息
ISBN:
(纸本)0769529097
When real-time embedded systems are constructed using conventional software architectures, it is often practically impossible to verify their timing properties. This paper describes a software architecture based on pre-run-time scheduling that greatly simplifies the task of verifying a real-time embedded system's timing properties.
This paper introduces a new approach to build application layer multicast overlay - Multiple Shared Trees. Multiple shared trees' approach makes tradeoffs between traditional source-based trees and single-shared t...
详细信息
ISBN:
(纸本)9780769534343
This paper introduces a new approach to build application layer multicast overlay - Multiple Shared Trees. Multiple shared trees' approach makes tradeoffs between traditional source-based trees and single-shared tree, and between transmission efficiency and protocol overheads. Based on this, we propose two protocols to build ALM overlay among end users and media-forwarding-gateways respectively. The latter references the design thought of Aggregated Multicast to share the multicast trees among groups.
This paper proposes a novel algorithm for load balancing in the parallel and distributedsystems. Firstly, almost all jobs are organized into the standard job combinations, each of which consists of one to four jobs. ...
详细信息
ISBN:
(纸本)0769529097
This paper proposes a novel algorithm for load balancing in the parallel and distributedsystems. Firstly, almost all jobs are organized into the standard job combinations, each of which consists of one to four jobs. Then they are assigned to the machines according to the assignment algorithm for job combinations, which is a special integer partition algorithm. The maximum error between the optimal algorithm and our algorithm is estimated. The experimental results show that our algorithm performs efficiently.
We present an MPI + OpenACC implementation of the kernel-independent barycentric Lagrange treecode (BLTC) for fast summation of particle interactions on GPUs. The distributed memory parallelization uses recursive coor...
详细信息
ISBN:
(数字)9781728174457
ISBN:
(纸本)9781728174457
We present an MPI + OpenACC implementation of the kernel-independent barycentric Lagrange treecode (BLTC) for fast summation of particle interactions on GPUs. The distributed memory parallelization uses recursive coordinate bisection for domain decomposition and MPI remote memory access to build locally essential trees on each rank. The particle interactions are organized into target batch/source cluster interactions which efficiently map onto the GPU;target batching provides an outer level of parallelism, while the direct sum form of the barycentric particle-cluster approximation provides an inner level of parallelism. The GPU-accelerated BLTC performance is demonstrated on several test cases up to 1 billion particles interacting via the Coulomb potential and Yukawa potential.
暂无评论