Action II(3) — Software Engineering UNified MEmory Computing (SEUNMEC)

Project Code: 2166-01/01-01
Funding programme: Action 2 - Support to Postdoctoral Researchers
Funding Agency: Athens University of Economics and Business
Project type: RTD
Starting date: 2018-01-01
Ending date: 2018-12-31
Total budget: 13,000 €


Scientific coordinator: Diomidis Spinellis

Description

In many situations the conventional operating system architecture imposes a heavy performance cost on common data access operations, such as reading and writing data, in exchange for security. One significant cause is that it is based on design aspects established decades ago when the landscape featured scarce and weak hardware resources, and limited virtual address space size for accommodating data. Nowadays, non-volatile, sizable storage devices with access speeds comparable to RAM hardware become increasingly affordable. In addition, typical computer systems ship with a 64-bit address space, which is very spacious. Although crucial factors have changed, the conventional operating system architecture remains eminent in today’s systems. The goal of the proposed project is to deliver architecture principles and software that significantly speed up tasks that rely extensively on operating system support, such as software builds and file processing, without compromising security. The goal is timely with innovations in memory hardware, which we will leverage to boost the efficiency of our approach. The proposed architecture and associated software, MemOS, will provide users with a standard command line interface, that is a shell, such as bash, through which they will be able to use their favorite build tools and other utilities that ship with or are provided for Unix-based operating systems; just as they do now.

Publications

PhD Theses

    • Marios Fragkoulis. Technologies for Main Memory Data Analysis. PhD thesis, Athens University of Economics and Business, Athens, Greece, 2017.

Journal Articles

    • Marios Fragkoulis, Diomidis Spinellis, and Panos Louridas. Live interactive queries to a software application's memory profile. IET Software, 13(4):241–248, August 2019.