Dartmouth College, Computer Science Department




Dartmouth College has a long tradition of excellence in computing. In the early 1960s, Dartmouth became one of the first institutions to make computers easily available to every student and faculty member, using a two-pronged approach that was revolutionary for the time. First, Dartmouth pioneered the use of the time-sharing operating system, thus providing interactive access to a large number of users. Second, two members of the Mathematics Department, John Kemeny and Thomas Kurtz, developed the BASIC programming language, a tool that made computing simple and easy for unsophisticated users.

The College nurtured the idea of "computers for everyone" throughout the 60s and 70s, led by the Mathematics Department and the Kiewit Computation Center. Mathematics courses included demonstrations by computer and occasional programming assignments. Every undergraduate was required to know enough to be able to log on to the time-sharing system and write some simple programs.

The growth of Computer Science, that is to say the study of the ideas underlying computation, was informal at first. Dartmouth granted a Ph.D. for a dissertation about programming language implementation in 1968, through the Mathematics Department. Throughout the early 1970s, the College introduced basic undergraduate courses in programming and data structures, compilers, data bases, and operating systems, taught by the Mathematics Department, and in digital logic and hardware design, taught by the Thayer School of Engineering. Graduate students could write computer science dissertations in both these academic organizations.

By 1977, undergraduates could major in mathematics with "concentration in computer science." In 1979, Dartmouth created the undergraduate major in Computer Science, to be administered by the Mathematics Department.

Dartmouth entered the 1980s with a healthy undergraduate computer science major in the Mathematics Department and hardware opportunities in the Thayer School of Engineering. There were occasional advanced seminars given by faculty from the Mathematics Department and the Thayer School. There was Ph.D. granting authority in Mathematics and the Thayer School, but there was no formal graduate program devoted to computer science.

From 1980 to 1988, the Computer & Information Science program (later Computer & Information Systems) granted 120 M. S. degrees. The program combined studies in information systems and business, and was independent of any department.

In 1983, recognizing the growing importance and influence of the computer scientists on its faculty, the Mathematics Department changed its name to the Department of Mathematics and Computer Science. Two years later, the Ph.D. program in Computer Science was approved by the Trustees of Dartmouth College and admitted its first students in the Fall term of 1986. In 1993, the computer science faculty moved to the wonderful new Sudikoff Laboratory for Computer Science, and in 1994, an independent Department of Computer Science was established. The new department is assuming administrative responsibility for the Ph.D. Program.

Today, Dartmouth's computer scientists are particularly proud and excited to have been a part of the construction of a new computer science building, the recent formation of the Department of Computer Science, and the continuing growth and strength of their curricular and research programs.

Within the Department of Computer Science at Dartmouth College there are several research laboratories and a machine room, which includes equipment from Aspen Systems that is dedicated to specific research projects:

  • For general department use is a 32 node 64 processor (Dual Xeon 2.8 GHz) Aspen Systems Linux Beowulf cluster with: 192 Gigabytes of RAM (6 GB per node) A 500 GB SCSI Ultra320 RAID array on the head node Ultra320 SCSI drives on each node (totaling 1152 GB) A Myrinet fiber network as well as a 1000 Base-TX copper Ethernet network MPI/PRO
  • Javed Aslam's Lab has a 12 node 24 processor (Dual Athlon MP 2000+) Aspen Systems Linux Beowulf cluster with 20 GB of RAM, 200 GB NAS, and a 1000 Base-TX copper Ethernet network.
  • Bruce Donald's lab has SGI workstations, Dual-Boot Linux-NT Dual-Processor Pentium workstations, assorted other Unix workstations, a Phantom force-feedback haptic device, an Ascension wireless motion-capture system, small RWI robots, and MEMS devices. The lab also has a 12 node 24 processor (Dual Athlon MP 2000+) Aspen Systems Linux Beowulf cluster with 36 GB of RAM, SCSI drives on half the nodes, a SCSI Ultra320 200 GB RAID Array on the head node, and a 1000 Base-TX copper Ethernet network.


“ It was hard to make a decision about which cluster vendor to go with, but we are very happy we went with Aspen Systems. To date, we have purchased 3 clusters (2x12nodes and 1x32 nodes) from Aspen Systems and have been satisfied with performance, price and support. The latest cluster we bought was especially fun to work with them on, as it's a really sweet machine. One thing that stands out about Aspen Systems compared to other vendors is that I have encountered a high level of friendliness and openness. I am sure we will continue to have a good relationship with Aspen Systems. We use the clusters for various research, from IR to parallel-IO to comp-bio to image processing to machine learning and robotics, among others. Everybody at Dartmouth College has been very happy with the clusters. We love it when experiments that used to take weeks per iteration now can be done in days or even hours. Thanks Aspen! ”
Tim Tregubov, SysAdmin, Computer Science Department, Dartmouth College