By now, most people have heard of “cloud computing” as a way to provide software and data to end-users.  Some companies have already deployed a cloud as their technology solution, while others are still evaluating whether it is a feasible paradigm based on their security and reliability needs.  Regardless of where one stands with respect to their consideration and use of the cloud, there are some specific e-discovery challenges associated with preservation and collection of cloud data that should be discussed.

Key components to the preservation process include the identification of potentially relevant information locations and the suspension of any auto-deletion processes that may destroy non-duplicative sources of ESI.  In a cloud, a user can access systems and information from anywhere and from any device.  While this is a great benefit of the cloud model, it can be quite challenging to nail down application specific locations of user generated ESI.  Some considerations for cloud preservation include:

• The infrastructure of a cloud environment can often be thought of as a “black box”.  For those cloud-based systems that are repeatedly relevant to litigation, a more detailed understanding of the back-end infrastructure is key and may even shed light on previously unknown locations of potentially relevant data.
• Cloud computing is based on the dynamic “on-demand’” allocation of resources based on user need.  These resources are then de-allocated as the need shrinks.  As such, it’s critical to ensure that potentially relevant user generated ESI is not also de-allocated solely because the need for computing power has decreased.
• If an application allows a user to generate ESI, consider enabling functionality that would create a preservation copy of that ESI on a by-user basis.

Collection of potentially relevant ESI typically requires forensically sound copies in preparation for the processing phase of the e-discovery life cycle.  In a “cloud computing” model, one may not always know the actual location of the data.  Some considerations for collection in the cloud include:

• Data transfer speeds over the Internet for large volumes of data can be slow.  If a collection endeavor requires forensically sound copies of server email for 40 custodians, and each custodian email is at least 1 GB in size, there is potential for data transfer of more than 40 GB of data from the cloud.  One should therefore consider the time it can take for large volumes of data collection when implementing a cloud model.
• Depending on your cloud service provider, the data may not physically reside in the same country as your company.  As such, the ESI may be subject to specific data privacy regulations that need to be considered prior to performing any type of collection.  This can be particularly important for US corporate data that physically resides in the EU, as their data privacy laws and regulations are much more stringent that US data privacy laws.

In my next blog posting I will expand on e-discovery in the cloud.

About Juilan Ackert

More ...