Case Studies

Introduction

Considering real life File system structures, we may also consider some Case Studies to get us through from issues to analysis, problem solving and developing solutions.

In our sample case studies, from small to big,

• Environmental Surveyor,
• IT company
• Researcher
• Home Office
• Solicitors
• Construction company
• Student MSc or PhD
• Consultant

Of the above companies we analysed in this mix, there is over 20 Terra bytes of data between them and not including databases and website (Intranet or Internet) data. We have not included ARCHIVES or various Backup data, on and offline, which would add another 30 terabytes on top of overall digital space. There are other legal issues to consider with some records that need to be preserved for at least 7 years in some cases. Other digital assets needed to be preserved indefinitely.

Most of the above companies used Windows as their base filing system or user Desktop and for this matter, file system format was mostly NTSF. While only 2 used a mixture of Linux and Windows. It would have been useful to consider a Graphics or Video editing company and we assumed this would be been based on Apple OS. Apple-OS is very similar to Linux/Unix in base file systems.

The cases covered a data mix which did not adopt cross platform replications or synchronisation. However, we noticed some did adopted a Google Drive, Dropbox, Browser replications and, collaboration, for both internal and external team mixes. This presented a few issues of redundant files and also orphan files. Some issues to consider in the future design.

As DAVit, specialised in data recovery, there are many other layers of deleted files on Media. So we advise to take account of all the hard and soft level digital space in a filing system. In many data recovery cases, the hidden layers of data helps to achieve a high level file recovery performance from disk failures or a disaster scenario. In some cases, with DVDs and hard disks, where there was serious physical destruction its still possible to achieve over 80% data recovery. This project is not about Disaster or Data recovery but should be implemented in the fileme Protocol and with emphasis not just on data recovery performance but also on the speed of recovery from a disaster scenario from small one person operation to a large company mix.