Posts Tagged ‘encapsulate
Purpose of TC Synchronization and Channel Coding
The purpose of the Synchronization and Channel Coding Sublayer and the associated Physical Layer Operations Procedures at the sending end is to:
- encode the data units received from the sublayer above to provide a forward error detection and correction capability, which gives a high degree of protection against errors or corruptions that occur during transmission through the space link;
- encapsulate the data units so that the start and end can be detected by the receiving end;
- enable the receiving end to resolve the data ambiguity (sense of ‘1’ and ‘0’) in the received symbol stream;
- control the transmission of an acquisition bit pattern, which enables the receiver to acquire bit synchronization;
- ensure there are sufficient bit transitions in the transmitted bit stream so that the receiver can maintain bit synchronization during the reception of a data unit;
- control the optional transmission of an idle bit pattern, which enables the receiver to maintain bit synchronization between data units.
The purpose of the Synchronization and Channel Coding Sublayer at the receiving end is to:
- detect the start and end of each data unit in the received symbol stream;
- resolve, if necessary, the data ambiguity (sense of ‘1’ and ‘0’) in the received symbol stream;
- decode each data unit and optionally correct the errors detected.
Data Grid Design
The following principles derive from the fact that data grid applications must frequently operate in wide area, multi-institutional, heterogeneous environments, in which we cannot typically assume spatial or temporal uniformity of behavior or policy.
Mechanism neutrality : The data grid architecture is designed to be as independent as possible ofthe low-level mechanisms used to store data, store meta data, transfer data, and so forth. This goal is achieved by de ning data access, third-party data mover, catalog access, and other interfaces that encapsulate peculiarities of specifi c storage systems, catalogs, data transfer algorithms.
Policy neutrality : The data grid architecture is structured so that, as far as possible, design decisions with signifi cant performance implications are exposed to the user, rather than encapsulated in “black box” implementations. Thus, while data movement and replica cataloging are provided as basic operations, replication policies are implemented via higher-level procedures, for which defaults are provided but that can easily be substituted with application-specifi c code.
Compatibility with Grid infrastructure : We attempt to overcome the difficulties of wide area, multi-institutional operation by exploiting underlying Grid infrastructure that provides basic services such as authentication, resource management, and information. To this end, we structure the data grid architecture so that more specialized data grid tools are compatible with lower-level Grid mechanisms. This approach also simpli es the implementation of strategies that integrate, for example, storage and computation.
Uniformity of information infrastructure : As in the underlying Grid, uniform and convenient access to information about resource structure and state is emphasized as a means of enabling run time adaptation to system conditions. In practice, this means that we use the same data model and interface to access the data grid’s metadata, replica, and instance catalogs as are used in the underlying Grid information infrastructure.
These four principles lead us to develop a layered architecture (Figure 1), in which the lowest layers provide high-performance access to an orthogonal set of basic mechanisms, but do not enforce specifi c usage policies. For example, we defi ne high-speed data movement functions with rich error interfaces as a low-level mechanism, but do not encode within these functions how to respond to storage system failure. Rather, such policies are implemented in higher layers of the architecture, which build on the mechanisms provided by the basic components.
Figure 1: Major components and structure of the data grid architecture
This approach is motivated by the observation that achieving high performance in speci c applications often requires that an implementation exploit domain-speci c or application-speci c knowledge. In data grids, as in other Grid systems, this focus on simple, policy-independent mechanisms will encourage and enable broad deployment without limiting the range of applications that can be implemented. By limiting application speci c behaviors to the upper layers of the architecture, we can promote reuse of the basic mechanisms while delivering high-performance and specialized capabilities to the end user and application.