Instructors

Prof. Dr. Tilmann Rabl, Lawrence Benson

Description

The ever-increasing volume of data that is produced today poses numerous challenges on how to efficiently manage it. One of these challenges is how to efficiently store and retrieve such data volumes. For many years, the memory hierarchy was clearly split into fast and expensive DRAM vs. slower but cheaper SSDs (or HDDs). The main goal was to squeeze as much hot data as possible into limited DRAM and only access disks for cold data. This came at the price of either giving up persistence for in-memory data or creating sophisticated mechanisms to store it persistently on slower disks. However, newly available non-volatile memory (NVM) is a class of storage technologies that provides fast byte-addressable storage (like DRAM) in much higher capacities than DRAM with persistence (like SSDs). Thus, NVM is a promising alternative to existing storage solutions. In this course we want to look at the latest developments in NVM-research and develop our own ideas on how to efficiently integrate NVM into the storage landscape.

Structure

Project

This course will be structured around group research projects of the students' choice in the field of NVM and related storage technologies. Depending on their interests, students will work in groups of 2-4 and come up with their own project ideas. The students should implement their ideas and evalute them. At the end of the course, the students should hand in a written report on their project. Ideally, we aim to publish the results at relevant conferences in this field.

Paper Discussions

To develop a solid understanding of current state-of-the-art around NVM, we want to discuss current research papers on a regular basis (i.e., weekly or bi-weekly). Each student should prepare such a discussion session and lead it. This involves studying the paper in detail, briefly presenting it, preparing potential discussion topics, and moderating the following discussion. 

Grading

The final grade will be made up of 50% project + report, 20% final presentation, 20% paper discussion, 10% active participation in all sessions.

Announcements

• The course will be managed via HPI Moodle. This is where we will announce things and share materials. 
• The course is limited to a maximum of 6 students.
• Kick-off Meeting, Wednesday, 4th November 13:30 in Zoom
• Due to space limitations and late Corona-related registration, please send me an email (lawrence.benson(at)hpi.de) if you want to participate in this course by Friday, 6th November 23:59. You do not need to send us this mail before the first event, so you can come and decide if you want to take this course after the intro. There is no first come, first served.

Schedule

The course will take place Wednesdays at 13:30. Due to the current Corona development, the course will kick-off virtually. We will evaluate if in-person meetings are possible once the situation improves. The meetings will be weekly (most weeks) to discuss a new paper and the individual teams' progress.