WebWinds Client, Server and Distributed Computing Architecture

This release of WebWinds features a distributed computing capability that allows data access via a computer, independent of the user's client machine, that is running WebWinds server software. If this server has access to data files, then any WebWinds client may access all or part of that data without the need for explicitly transferring the data to the client. In the example shown below, data files are accessible via a central machine (e.g. a Unix workstation).





Clients on the intranet (e.g. PC's, Mac's) access the data in a transparent way after specifying the server's name or IP address. More than just a file finder, the server actually devotes processor cycles to reading, subsetting, subsampling and/or interpolating data. Clients may also access other servers elsewhere on the Internet.
 

Server Frequently Asked Questions (FAQ)

  1. Do I need special software to run the server? No. All that is needed is the WebWinds server software (including Java) which is easy to install on all platforms.
  2. Can other (non-WebWinds) applications make use of the server? Not at this time. Future development would allow other Java applications to access the WebWinds server.
  3. Can the server be a Mac or PC?Yes.
  4. How do the clients find the files they are interested in? The server produces a FileFinder, that is exactly the same as the one used for the client. With this and knowledge about the file and directory structure on the server, the user migrates to the file of interest.
  5. I'm worried about security. How do I know that hackers won't break into the server? Although there are no guarantees, this is extremely unlikely as described in the server documentation. There are protections at many different levels, making this much more secure than Windows or even Unix operating systems.
  6. Can I limit access to the server? At this time, you can limit which files/directories can be read but not who can 'login' to the server. However unless you tell people about the existence of your server nobody will know it is there.
  7. You call this 'distributed computing'. Why should I care about this? Let's say you have a meager PC with 64 MB of memory on your desktop but have a need to use very large data sets. Let's also suppose that the machine with access to those data sets is a powerfull (256 MB of memory, fast processor) Unix machine. Transfering huge files to your PC  so that you can look at a small portion of the data may be unworkable. With the server, it can be done painlessly.

WebWinds Home / Oct 5, 2001