Basics of Data Management
Introduction
Data Management (DM) allows a design team to manage and track the creation, editing, and archiving of all data and information related to a project. The data is kept in a central repository and the data management tool will control the check-in and check-out of files to a multi-user community of designers, insuring that only one user has write access to the data at a given time. As users check-out, edit, and check-in data, the data management tool assigns a version number to each check-in, thus keeping a history of the changes made to the data which is available for roll-back if deemed necessary.
Storage of Managed Data
There are two methods that are commonly used by data management tools for storing and managing data:
Copy Method: Tools that utilize the Copy method make copies of the data. Each version is a new copy in the data repository. For example, if there are 4 versions, then there are 4 files stored in the repository. As a result, this method consumes the largest amount of disk space.
Delta Method: Tools that utilize the Delta method store the original file and the differences between the versions. For each managed file, there is only one file in the data repository, and each version is recorded as a series of changes, often stored at the end of the file itself.
Regardless of the method used for versioning, there is one basic requirement that a data management tool must meet: when data is checked out, the managed data must be in an unaltered form. What this means is that when the data is checked out, it must appear to be a normal file usable by the tool. Any tool capable of operating on the data in an unmanaged state must be able to operate on that data in a managed but checked out state.
The LDM Facility
LDM stands for Laker Data Management. The LDM facility is a set of Tcl commands that can be used to perform DM operations such as check in and check out of files. LDM facilitates the access of data stored in the Laker DB structure, including properly handling co-managed sets and other data that must be accessed in a synchronized fashion. Third-party DM tool providers communicate with the LDM through the Tcl commands. As a result, the LDM facility is transparent to end users. Data management may be used with both the Laker DB and Laker OA.
Workareas
Workareas are directories set up for users to check out, modify, and check in files. In a workarea, the files are read-only copies unless a file has been checked out for edit, then that particular file will have write privileges. There are 2 models used for workareas plus a combination of the 2. They are:
Shared Workareas - There is one project workarea that all users share, also called the Sandbox Model.
Individual Workareas - Each user has their own individual project workarea.
Combination - A combination of both may be used.
Shared Workareas
In the shared workarea method, there is one project workarea and all users work in this same workarea. There are advantages and disadvantages to this model. One of the advantages to this model is that it consumes less disk space. Having only one workarea means only one copy. A user may check out a file for edit, and perform edits. Once the user saves the edits, everyone else working in that workarea will see those edits, even if the file has not been checked in to the repository. This can be both good and bad. The advantage is that the users have immediate access to the modified file and may test the edits before the modifier checks the file into the repository. Another advantage is that if the file has been checked in, all users will have access to this latest revision without having to perform a workarea update.
The disadvantage is that the file may have had an intermediate save without any verification. This file may not interact properly or as expected with the other files in the workarea. This could cause problems for users dependent on that file being correct and verified. A simple example of this is if pins are added or removed from a cell, and higher level cells instantiating that cell have not yet been updated to reflect the modification, the users could encounter problems or errors. The users are working in a dynamic environment. The shared workarea model is not widely used due to this disadvantage. It takes excellent communication between users to maintain a smooth flow when operating under the shared workarea model.
Individual Workareas
In the individual workarea method, each user works in a separate workarea. There are also advantages and disadvantages to this model. The advantage is that the users may work independently on various files without having to worry about dependent files being modified. They work in a stable environment. The normal process is to check out a file, perform edits, verify edits, and check the file back in. An updated read-only copy of the edited file will remain in the user's workarea. With the individual workarea method, users may choose when to update their workarea with newer revisions.
The disadvantage is that all other users must perform a workarea update to receive copies of any newly modified cells. Most DM tools have an email feature that will notify users when a file has been checked in to the system, but the users will have to perform the update. If multiple users are modifying files, they could still create conflicting edits, so communication and coordination is still important in this model.
Combination
A combination of shared and individual workareas may also be used. It will have a combination of the advantages and disadvantages of both the shared and individual workarea models, and communication between designers is very important to maintain a workable environment.
Summary
Data Management is a very useful tool for keeping track of revisions. There are options for setting up your environment that should to be reviewed. The specific DM tool provider may have recommendations for using their tool. These should also be considered when setting up the DM environment.
