Ensuring Data Consistency in Large Network Systems
نویسنده
چکیده
Data management in network systems is different from that in traditional data systems. It is important to ensure data consistency not only between network management systems, but also between network elements, which may not support database protocols such as two-phase commit, as well as between network management systems and network elements. The problem is nontrivial, especially when data are updated unilaterally from network elements without notibing network management systems, as network management systems and network elements operate independently and autonomously. This paper presents a technique that maintains data consistency in large network systems. The technique consists of a suite of protocols that allows creation, deletion, and modification, as well as downloading of data to network elements and verification of data consistency between network management systems and network elements. Together, these protocols ensure consistency when data are not updated directly from network elements, and otherwise detect and resolve inconsistencies as long as updated data have not been used by network elements for configuration.
منابع مشابه
Consensus in a Box: Inexpensive Coordination in Hardware
Consensus mechanisms for ensuring consistency are some of the most expensive operations in managing large amounts of data. Often, there is a trade off that involves reducing the coordination overhead at the price of accepting possible data loss or inconsistencies. As the demand for more efficient data centers increases, it is important to provide better ways of ensuring consistency without affe...
متن کاملA Replica Control Method for Improving Availability for Read-only Transactionst
Data replication is often considered in distributed database systems to enhance availability and performance. The benefit of data replication, howevel; can only be realized at the cost of maintaining the consistency of data. In part i d a l ; network partition failures make it more dificult to achieve high data availability while ensuring strong correctness criteria such as I-copy serializabili...
متن کاملA Replica Control Method for improving Availability for Read-only Transactions
Data replication is often considered in distributed database systems to enhance availability and performance. The benefit of data replication, however, can only be realized at the cost of maintaining the consistency of data. In particular, network partition failures make it more difficult to achieve high data availability while ensuring strong correctness criteria such as 1-copy serializability...
متن کاملDistributed GraphLab: A Framework for Machine Learning and Data Mining in the Cloud
While high-level data parallel frameworks, like MapReduce, simplify the design and implementation of large-scale data processing systems, they do not naturally or efficiently support many important data mining and machine learning algorithms and can lead to inefficient learning systems. To help fill this critical void, we introduced the GraphLab abstraction which naturally expresses asynchronou...
متن کاملDistributed GraphLab: A Framework for Machine Learning in the Cloud
While high-level data parallel frameworks, like MapReduce, simplify the design and implementation of large-scale data processing systems, they do not naturally or efficiently support many important data mining and machine learning algorithms and can lead to inefficient learning systems. To help fill this critical void, we introduced the GraphLab abstraction which naturally expresses asynchronou...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2001