DRAMaOS: DRAM-aware Operating Systems (DFG: LO 1719/10-1 and JU 3158/1-1)

DRAMaOS is a joint project with the CE groupt at Universität Würzburg (Co-PI: Matthias Jung) and part of the DFG Priority Program "Disruptive Memory Technologies" (SPP 2377, Speaker: Olaf Spinczyk).

The computer’s main memory is arguably the most fundamental kind of memory the operating system (OS) has to manage, with the memory page being the defacto entity for all usages of main memory.

In the hardware, main memory is mostly provided by DRAM, for which advances in density have made it possible to build systems with TiB of main memory. Internally, modern DRAM subsystems employ complex topologies with multiple channels or pseudo-channels distributing the memory and its accesses across ranks, bank groups, banks and rows in order to improve nonfunctional properties, such as access latency and throughput, energy consumption, robustness and security. All these are influenced by the actual access and allocation patterns at run time. Upcoming standards, such as LPDDR6, DDR6 or HBM4, provide even more features to trade available memory, throughput, energy and security, while new trends, such as CXL.mem and processing-in-memory (PIM) add completely new dimensions to the DRAM hierarchy.

So far, most of this is kept hidden from the OS and its memory-usage strategies behind the memory controller (MC). However, there is significant potential for optimization if the OS becomes more DRAM-aware – and the MC more OS-aware. Moreover, with a programmable “active” MC, the OS could even offload specific memory maintenance tasks, such as page zeroing or copying. Exploiting such features would not only require a much more accessible interface to the MC, but also the OS to manage main memory for many different granularities and usage patters. Because all memory is not equal.

In DRAMaOS, we investigate the OS–MC interface from both sides. We explore the potential of a OS–MC codesign approach to improve nonfunctional properties (such as energy consumption) of DRAM-intensive systems by adaptable DRAM-aware OS strategies and OS-aware MC functionality.

People

Illia Ostapyshyn (M.Sc.)

Nils Wilbert

Prof. Dr.Ing. Matthias Jung

Prof. Dr.-Ing. habil. Daniel Lohmann

Latest News

2025-09-25 Illia Ostapyshyn Receives GI-FGBS Award for Best Master Thesis

Illia Ostapyshyn receives the award for best master thesis in the field of operating systems. The award is granted annually by the SIG on Operating Systems of the German Computer Association (GI Fachgruppe Betriebssysteme) solely on the base of scientific excellence. It includes a price money of 500 €. Congrats, Illia!

Recent DRAM standards provide increasing support for energy saving modes. However, current OS are not able to make use of them because of the way they allocate and use physical memory. In his thesis Enhancing Energy Efficiency with Advanced DRAM Management in Linux Illia designs, implements and evaluates a new concept for DRAM-aware memory allocation. This excellent work is now continued within the DRAMaOS project.

2025-09-08 New DFG Grant: DRAM-aware OS (SPP 2377, Phase 2)

DRAMaOS: DRAM-aware Operating Systems (DFG: LO 1719/10-1 and JU 3158/1-1): In DRAMaOS, we examine new approaches to improve nonfunctional properites (such as energy consumption or security) of DRAM-intensive systems by adaptable DRAM-aware OS strategies and OS-aware memory-controller design.

DRAMaOS is a new project within the DFG Priority Program "Disruptive Memory Technologies" (SPP 2377). DFG is supporting us for three years with two positions for doctoral researchers (E13, one at SRA), two positions for student researchers (one at SRA), and some additional lab equipment. DRAMaOS is a cooperation project with the Computer Engineering group of Matthias Jung.

Publications

DIMES Workshop Best Paper Award The New Costs of Physical Memory Fragmentation

Alexander Halbuer, Illia Ostapyshyn, Lukas Steiner, Lars Wrenger, Matthias Jung, Christian Dietrich, Daniel LohmannProceedings of the 2nd Workshop on Disruptive Memory Systems (SOSP-DIMES '24)Association for Computing Machinery2024Best Paper Award.
PDF Details 10.1145/3698783.3699378 [BibTex]

Theses

Currently Running

To Drop or Not To Drop: Identifying Unused File Cache in Linux

Develop a mechanism to identify eviction candidates among unmapped file-cache pages in the Linux kernel.

Typ: Masterarbeit
Status: laufend
Supervisors: Illia Ostapyshyn
Daniel Lohmann

Leveraging Highly-Volatile Pages to Improve Memory Compaction Efficiency in Linux

Studies report that the majority of allocated pages become free within a timespan of seconds. Can a lazy compaction algorithm use this fact to achieve compact memory arrangement with less effort?

Typ: Bachelor-/Masterarbeit
Status: reserviert
Supervisors: Illia Ostapyshyn
Daniel Lohmann

Finished Theses

System-Independent Cost Model Derivation for Energy-Aware Compaction

Effective memory compaction and reclamation with the goal of saving energy requires precise prediction about its costs and benefits. In this thesis, statistical methods shall be evaluated to automatically derive a cost model for energy-aware compaction.

Typ: Bachelorarbeit
Status: abgeschlossen
Supervisors: Illia Ostapyshyn
Daniel Lohmann
Bearbeiter: Anastasiia Martynkova (abgegeben: 14. Jul 2025)

Enhancing Energy Efficiency with Advanced DRAM Management in Linux

Modern computer systems face a significant challenge in managing physical memory efficiently, especially in the context of energy consumption. Specifically, the DRAM (Dynamic Random-Access Memory) components in these systems consume a considerable amount of energy due to their constant refresh requirements. [PDF]

Typ: Masterarbeit
Status: abgeschlossen
Supervisors: Alexander Halbuer
Christian Dietrich
Daniel Lohmann
Bearbeiter: Illia Ostapyshyn (abgegeben: 01. Nov 2024)