Back to search

EUROSTARS-EUROSTARS

E!12494 Immortal Virtual Machine - solving the problem of file format and infrastructure obsolescence

Alternative title: Immortal Virtual Machine - Løser problemet med utdaterte filformat og infrastruktur

Awarded: NOK 6.0 mill.

Project Manager:

Project Number:

291182

Project Period:

2018 - 2021

Funding received from:

Organisation:

Location:

Partner countries:

The goal of the Immortal Virtual Machine (iVM) project is to build a comprehensive preservation solution that ensures access to current digital data in the future (500+ years). We have developed a Virtual Machine (VM) to reconstruct data in the future, it is integrated with Piql Services. A virtual machine is a virtualised computer, that can run and deploy programs. The preservation virtual machine is specifically designed for archiving purposes. It comes with a minimalistic implementation guide, designed for the distant future, where contemporary computers and SW no longer exists. This machine will simplify the process for obtaining access to all data stored on piqlFilm. It will also ensure that file formats stored on piqlFilm are rendered and interpreted correctly. The virtual machine implements self-executing file format decoders that decode the file formats and produce consistent output to input/output devices (i.e. an image appears the same regardless of the device it is viewed on). The VM automates a large part of the manual recovery of files from piqlFilm. Written in human readable text on piqlFilm, the virtual machine will decode and compile automatically and render the files recovered, which allows the user to open and access the information. This will have a significant impact for the user working to recover the information. The virtual machine renders preservation format files such as PDF/A, TIFF, WAV etc.

The virtual machine gives data "eternal life" The virtual machine implements self-executing file format decoders that decode the file formats and produce consistent output to input/output devices (i.e. an image appears the same regardless of the device it is viewed on). Now written in human readable text on piqlFilm, the virtual machine decodes and compiles automatically, rendering the files, allowing the user to open and access the information. This will have a significant impact for the user working to recover the information. The virtual machine renders preservation format files such as PDF/A, TIFF, WAV etc. Data security: Data on piqlFilm cannot be altered or overwritten, therefore ensuring that data cannot be manipulated, altered or deleted. Sustainability: piqlFilm can also be stored in the Arctic World Archive, where the natural climatic conditions allow storage with no electricity required.

The goal is to design, implement and demonstrate a ground-breaking Virtual Machine (VM) that future software developers can reconstruct on their infrastructure – for the purpose of viewing/rendering data that was stored in the past. To do this the developers will read a visual description and implement the VM by loading additional VM code, both stored on a 35mm and ultra-high resolution photosensitive film (piqlFilm). The VM is then able to retrieve and render content stored on the film in selected file formats. The concrete scientific, technical and commercial results to be achieved are listed below: Scientific 1. A simple and unambiguous description of a VM for content rendering, i.e. a VM that can be understood and implemented on future computing infrastructure by future developers (e.g. 500 years from now) with a minimal effort; 2. Algorithms to translate (cross-compile) a high-level programming language (C/C++) that include various file format functions into the simpler language of VM code; 3. Algorithms for optimizing VM code to ensure efficient execution of the VM. Technical 1. A reference implementation of the VM and additional software for reading data from film and rendering content on output devices. This will be used to demonstrate and validate that the technology is working; 2. A Compilation System targeting the VM, a code generator for VM code, and an optimizer for VM code; the system will be used to translate existing C/C++ source code for processing file formats to VM code. 3. Development and optimisation of the complete data preservation workflow that automates and quality controls data flow from client through the preservation system and onto the long-term storage medium. Commercial 1. Test highly relevant use cases from prominent users like UNESCO, the National Museum of Norway and the Vatican Library to demonstrate iVM technology in a relevant environment (TRL7); 2. Provide a commercial plan to realise about €633m of VM Services sales 5 years

Funding scheme:

EUROSTARS-EUROSTARS