Research Position: High Performance I/O

The Max-Planck-Institut für Informatik is offering a PhD or Postdoc position in context of an international project on algorithmic support for high-performance I/O.

[disks] WIDTH=142 Connecting many hard disks to a high performance computer in principle allows to efficiently process huge amounts of data. We want to make it easy to program such systems. Details like data placement, load balancing and fault tolerance should be hidden in lower levels of a software library. On the upper levels frequently needed algorithmic components such as sorting or search trees should be available. Our project is sufficiently flexible so that we could also tackle other interesing challenges such as

Innovative file servers
Parallel file systems
The Elephant-PRAM: Shared memory for large data sets
Simple high performance data base kernels

What we offer

you work in an international cooperation of renown groups in the area of algorithms and software libraries
Generous travel budget
No teaching obligations
Internationally competitive payment

What we want

A MSc in computer science or equivalent
Basic competence in algorithms and data structures
Good programming skills, e.g., in C/C++
Interest in programming close to the operating system
Fluency in English (no German required)

For more information contact Peter Sanders, sanders@mpi-sb.mpg.de. Applications (including a covering letter indicating your motivation, a short curriculum vitae, transcripts documenting your studies, names of 2-3 people whom we may ask for letters of recommendation) should be sent as soon as possible to

Peter Sanders, Dr. rer. nat.
Max-Planck-Institut für Informatik
Stuhlsatzenhausweg 85
66123 Saarbrücken
Germany

Email: sanders@mpi-sb.mpg.de
Phone: ++49 681-9325 115
Fax: ++49 681-9325 199
WWW: http://www.mpi-sb.mpg.de/~sanders/

The Max-Planck-Society intends to increase the number of women in those areas where they are underrepresented. Women are therefore explicitly encouraged to apply.

Some literature

Andreas Crauser. LEDA-SM External Memory Algorithms and Data Structures in Theory and Practice. PhD thesis, Universität des Saarlandes and MPII, 2001.
D. A. Hutchinson, P. Sanders, and J. S. Vitter. Duality Between Prefetching and Queued Writing with Parallel Disks. In 9th European Symposium on Algorithms (ESA), LNCS. Springer©, 2001. to appear.
P. Sanders. Reconciling Simplicity and Realism in Parallel Disk Models. In 12th ACM-SIAM Symposium on Discrete Algorithms, pages 67-76, Washington DC, 2001.
P. Sanders. Asynchronous Scheduling of Redundant Disk Array. In 12th ACM Symposium on Parallel Algorithms and Architectures, pages 98-98, 2000.
P. Sanders, S. Egner, and J. Korst. Fast Concurrent Access to Parallel Disks. In 11th ACM-SIAM Symposium on Discrete Algorithms, pages 849-858, 2000. full paper.