Department Databases at Intel

At Intel there are many databases that we use to do business. Most of the databases that we use are very complicated and require a lot of training and proficiency to surmount the program. In the clean room at Intel Corporation, we use a specific type of database to keep track of each and every microchip that we make. Since we make millions and millions of microchips and it takes many different machines, processes, and people to make these chips, the database has to be able to handle all of this information. This database that we use is called Work stream.

Work stream is such a complicated interactive database that is much different than any other that I’ve seen before. It is a UNIX based system that looks very old. It might look old, but the function is outstanding. In fact, it acts like it is its own operating system because it is so big and has many different commands to get you to certain pages. The input comes from two different places. One place is from the operators or technicians that make the microchips, and the other is from the tool sets (or equipment) themselves. These commands that are inputted take you to a particular page or query in the database where you can add different data about the microchips that we process, or even the tool sets that the lots of microchips process on. Every lot (or group of microchips or wafers) and every tool set has its own identification number that is inputted into the database for tracking. Everything that is done to any microchip and tool is documented. For instance, when we do preventive maintenance on the tool, we document everything that we do to it in Work stream to verify what we have done. This way, if something goes wrong, we can go back to the time of the repair in specific detail, almost to the second. It is the same way for the microchips as well, however, the tool set automatically updates the databases every time a certain lot processes on the tool. So, if needed, one could go back into the Work stream and find out exactly what time and process was used on a particular wafer or microchip because the tool updated the system automatically. There are also reasons for people to input data or information about the microchips in Work stream as well. Sometimes it is necessary to add input to a certain problem a microchip might have faced or have been damaged on a certain tool or process. Most of the time this is not needed until an excursion or something goes wrong that isn’t standard processing. All in all it is a great database to be working with and pretty much takes care of every aspect of data control in the entire clean room and the entire creation of a particular product.

Work stream is an old system and has been around since Intel Corporation started. Its appearance looks like a dos screen with the same type of black screen with white letters. This is one thing that needs to be changed I think because it is not user friendly in this day of age. Yes, it is a good system to use and that is why Intel stuck with it for so long. But the appearance of most modern databases is so much nicer and very pleasing to look at, making it more user friendly.

Speaking of not being user friendly, Work stream requires a massive amount of training for one to be proficient and to truly work their way around in the system. There are so many different commands that we use to take us to different places to where we can either