Euromug '96 - New projects - previous - next


The project goal is to implement a medium-to-high performance chemical information server and use it to deploy large chemical databases on the Internet.


Design and initial negotiations with data vendors are underway. Human resources and computer infrastructure could be in place mid-1997. Databases and pre-production software could be in-place by the end of 1997.



The Mjollnir project was spawned by an analysis of the sociological aspects of scientific information exchange by Howard Winant about 16 years ago (not only pre-Daylight, but pre-MedChem!) One of the conclusions of this analysis was that of the three sociologically-evolved methods of scientific information exchange (reviewed journal, library, forum), only the forum is suitable for use on a global network in a fully distributed manner. In an idealistic forum, participants must identify themselves, may freely (or at least equally) have access to all information, and may make contributions without prior review or censorship (including comments about other people's contributions). Most scientific forums take the form of conferences, but the forum has no requirement for immediacy, e.g., usenet is a good example of a forum.

The Thor database system was designed with the forum concept in mind, although it is not (yet) used this way. Universal structure-based indexing, constant time retrieval and strict ID-data distinction make Thor particularly suitable for use in a chemical information forum.

An early version of Mjollnir which implements (only) universal access of low-performance data retrieval via e-mail has been operational for several years. It is primarily used by academics with poor access to information. Although this system makes a lot of data available (the medchem, tsca, spresipreps, and wdi databases), its very low performance makes it immune to "database-raping" and limits its use to delivering data to desparately poor students and marketing databases to better-funded researchers (e.g., database demos).

The current Mjollnir project is aimed at raising the system's performance to a more usable level and introducing some forum features. Key design features include:

Given the realities of how most chemical information is collected (expensively) and disseminated (expensively and not that universally), the Mjollnir approach seems to be in everyone's benefit:


We expect to start with a Sun 4000 Enterprise which Sun has loaned us for server development. Enterprise machines are great for servers such as ours: the architecture scales up to 30x500 MHz CPUs, 30 GB RAM, and 6 TB disk. They claim that the new asynchronus memory manager can keep up with all those CPU cycles (the 500 MHz CPUs aren't shipping yet, we'll see...) As a bonus, they have very robust (fault-tolerant) features.

Physical environment

We are building an office with provisions for uninterruptible power and a T1 line which will serve as a stable environment for the service. Should be ready by mid-1997.  

Human resources

Two additional research office staff will come on board in early 1997: one primarily for research support, the other a network/java-interface guru. Both will spend some of their time on this project.

As always, we are looking for input and feedback from our users. If you have ideas about the role, design, or implementation of Mjollnir, let us know. Soon!

Daylight Chemical Information Systems, Inc.