Electrical computers and digital processing systems: support – Digital data processing system initialization or configuration
Reexamination Certificate
2000-11-20
2004-09-14
Lee, Thomas (Department: 2115)
Electrical computers and digital processing systems: support
Digital data processing system initialization or configuration
C713S002000, C709S222000
Reexamination Certificate
active
06792526
ABSTRACT:
BACKGROUND OF THE INVENTION
Field of the Invention
Topologically flexible and large, distributed systems (i.e. computer systems) require that the process of starting up (i.e. booting up) the individual components be coordinated when such systems are started up overall (initial commissioning or restarting of the overall system in order to neutralize hardware/software faults). In the following text, the term component has the following meaning: a processor with a software function or with a number of software functions, or a group of processors which each contain one, several or mixed software functions. The term starting up of a component refers to the loading of the software function or functions in the hardware part of the component, and the initialization and activation of the software function or functions. (It is necessary to distinguish from this the process of starting up software processes on an individual platform whose operating system is already in operation, so that process activity is feasible, as described, for example, in the document International Patent Disclosure WO 97/01815.
A coordinated starting up of the system is required, inter alia, since the components to be activated require software functions (code and data) which they need to start themselves up and to start up components downstream from them in the hierarchy. A starting-up process staggered in time and topologically during an overall start-up process is thus necessary, in contrast to local or other networks in which the network components register with their master/server, or attempt to do so until the master/server can be reached.
One example of large distributed systems is switching systems for broadband users (EWSX or, in EWSD, the SSNC). Different software functions are distributed between a number of processors in these systems. Furthermore, these systems are characterized by a distributed operating system and by a distributed database. Different component groups have direct or indirect relationships with other components in the system during an overall start-up process.
First, the customers require topologically highly different systems, that is to say the number of processors in the system and their linking structure may vary, and, second, the customers require functionally highly different systems, that is to say the distribution of the software functions between the various processors in the system depends on the customer requirements.
Furthermore, new components are constantly being developed, which are integrated into the existing system as development progresses.
What has been stated above results in the following requirements for the control software for an overall start-up process.
The control software (start-up control component) for coordination of the starting-up process (other terms used for “starting-up”: activation, booting up, starting) of the overall system should be configured such that it manages the process of starting up the system independently of changes to the architecture/topology.
In particular, the starting-up process should be configured to be independent of hardware so that, for example, the control software also copes with upgrading or removal of entire parts in the topology. This also results in greater stability and reliability on activation of the overall system.
In the following text, the previous implementation of the start-up process in the specific EWSD switching system from Siemens will be explained in more detail.
Even with EWSD, the number of LTGs, DLUs and their data (ports) are likewise taken from the database, which results in a certain amount of flexibility. However, the following factors must be taken into account:
a) the sequence of activation of the individual components is hard-coded (there are specific activation and synchronization routines for each component);
b) the subsequent activation time for the individual components is hard-coded, that is to say how much progress the already started component can make (satisfaction of the preconditions for starting subsequent components) before the next component group for activation can be started;
c) redundancy principles are hard-coded (message buffer/switching network); and
d) the handling of LTG types is different, and is hard-coded.
In the EWSD system, there are also a number of additional aspects (other preconditions), they include:
a) the EWSD has a limited number of processors (6 processors) and a rigid hierarchy of EWSD processors is not a system with a distributed database;
b) the EWSD has a common memory for data (common memory);
c) the EWSD is not a system with a distributed operating system; and
d) the EWSD has a rigid function split, that is to say no arbitrary distribution of software functions between the processors.
The system has the below listed disadvantages. If the EWSD is or needs to be upgraded, changes are required in the software which controls the activation (high change complexity: analysis, design, implementation, test . . . ).
With the way in which the system peripherals has so far been taken into operation with hard coding in the EWSD, the future customer requirements with regard to flexibility of switching systems, in particular with regard to the topology, can no longer be satisfied cost-effectively and quickly.
SUMMARY OF THE INVENTION
It is accordingly an object of the invention to provide a control system for controlling the starting up of a distributed system which overcomes the above-mentioned disadvantages of the prior art devices of this general type.
With the foregoing and other objects in view there is provided, in accordance with the invention, a control system for controlling a booting-up of a distributed system containing a number of components. The control system contains a control component for controlling a booting-up process of the distributed system; and a database connected to the control component and storing data related to each of the components of the distributed system. The data describe characteristics of the components in terms of control of the booting-up process. The control component is programmed to read the data relating to a boot-up time for each of the components from the database, and uses the data which have been read to control the booting-up process of the distributed system.
In accordance with an added feature of the invention, the control component controls the booting-up process of the distributed system in that it is programmed to use the data which have been read first to filter out the components which it will include in the booting-up process, and second to determine an activation sequence of the components that have been filtered-out.
In accordance with an additional feature of the invention, the data that have been read out contain information about at least one of subdivisions of component groups, processors, units, and platforms in hierarchies.
In accordance with another feature of the invention, the data that have been read out contain information about component statuses.
In accordance with a further feature of the invention, an interface through which an operator can use a command to select a scope of a component is provided.
In accordance with a further added feature of the invention, the control component is loaded with control software and, in order to identify a total failure before carrying out the booting-up process, the control software asks whether the total failure has occurred. In which case a total failure identification then uses the component statuses to determine whether an activation of an overall system to be carried out can provide a desired software function of a customer as an entity.
In accordance with a concomitant feature of the invention, each of the components of the distributed system is a processor having at least one software function or each of the components of the distributed system are a group of processors having at least one software function.
Other features which are considered as characteristic for the invention are set forth in the appended claims.
Although the invention is illustrated
Klein Holger
Niederau Bernd
Niessen Jürgen
Wich Klaus
Greenberg Laurence A.
Lee Thomas
Locher Ralph E.
Siemens Aktiengesellschaft
Stemer Werner H.
LandOfFree
Control system for controlling the starting up of a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Control system for controlling the starting up of a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Control system for controlling the starting up of a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3237732