| Back | Map | Glossary | Information |
First developed at CERN in Switzerland, the World Wide Web (WWW) has enjoyed enormous success and growth since the appearance of NCSA Mosaic and subsequent Web browsers (e.g. Netscape). The resulting exponential growth in network traffic has place unprecedented demands on Web servers around the world. Individual computers struggled and ultimately failed to keep up with the huge number of "hits."
During the two years following the release of NCSA Mosaic in 1993, the usage of the Web on the Internet grew from being the a fraction of one percent of total traffic (as measured in bytes or packets of data) to being the largest component. By April 1995, according to statistics from Merit Network, Inc., WWW data accounted for one out every three bytes of Internet traffic on the NSF backbone.
NCSA Web Server Stats
JPEG Images (36.7 KB) JPEG Images (36.7 KB)
NCSA's solution (and others followed suit) was to design a scalable server consisting of nine independent Hewlett Packard Apollo 9000 700 series workstations linked with a distributed file system (AFS).
NCSA Web Server System ("Round Robin" ANS)
Marshall Greenberg; access
Putting all of the information in the Andrew File System makes
it possible for each server to access all of the
information. Requests to the servers are passed to the different machines
in a "round-robin fashion." The result: a five-fold increase in possible
connections. Right now, the system is handling between 4 and 5 million connections per week.
JPEG Image (76.2 KB)
As the data storage and transfer needs of the Web continue to grow, particularly with the incorporation of video, the capabilities of information servers to handle the rising demand will be stretched to the limit. Research and development in both the public and private sectors are beginning to yield yet new software and hardware systems designed to handle rapidly the mounting volumes of data.
Return to the Parts of the Metacomputer