Skip to content

Vaults Overview

What are Vaults?

Vaults are similar to filesystems in that they provide a unified directory structure where folders, files, and datasets can be stored. All items in a vault (the folders, files, and datasets) are collectively referred to as "objects". All vault objects can be moved, copied, renamed, tagged, and assigned metadata.

  • Folders: Folders are objects that can contain other objects. The root of a vault is a folder represented as a single forward slash (/) in paths.
  • Files: File objects represent data in an underlying object storage system such as Amazon S3. The vault's "provider" defines where the underlying data is stored. Learn more about vault providers.
  • Datasets: Dataset objects represent data that is indexed for real-time querying and filtering. Learn more about datasets.

Vaults also have an advanced permission model that provides three different levels of access: read, write, and admin. Permissions can be set via the SolveBio UI (see Sharing Vaults).

For more information on permissions, see the SolveBio Help Center.

Vault Types

There are two types of vaults:

Personal Vaults

Every SolveBio user has a personal vault. Your personal vault is private and cannot be accessed by others or shared with others. To share contents of your personal vault you must move or copy the objects into a shared vault.

Shared Vaults

Shared vaults can be shared with individuals or groups within your account. Shared vaults are created within your account's "domain" and must have a unique name (within the domain). All vaults other than your personal vaults are shared vaults. At this time, vaults can only be shared within your account. One exception to this rule is SolveBio's public vault (solvebio:public) which is shared with all users from SolveBio's domain.

Learn more about using vaults →

Vault Providers

Vault Providers are in Beta

Please contact SolveBio for assistance using custom vault providers.

Vault Providers determine where and how the contents of a vault are stored. By default, vaults use SolveBio's built-in provider which stores files and datasets in SolveBio's Virtual Private Cloud (VPC) on Amazon Web Services. Files are encrypted and replicated on S3 and datasets are encrypted and replicated across EC2 instances in the same geographical region. Vaults can be created with different vault providers to access data from other services such as DNAnexus, S3, GCS buckets, or any other data platform.

Learn more about vault providers →

Paths and Filenames

All vault objects have a unique "full path". This path is composed of a few distinct elements:

+------+ +----+ +-----------------------+ +-------------+
(1)      (2)    (3)                       (4)
+-------------+ +---------------------------------------+
(5)             (6)

(1) - Account Domain
(2) - Vault Name
(3) - Object Parent Path
(4) - Object Filename
(5) - Vault Full Path
(6) - Object Path
(7) - Object Full Path

All objects in vaults have a "filename" (even folders and datasets!). Filenames can contain any character except forward slash "/" and can be up to 1000 characters in length. Filenames must be unique within a folder.