Active Archive Facility

Fermilab offers researchers access to remote storage of scientific data through the Strategic Partnership Project (SPP) mechanism.

Fermilab's world-class scientific data management capabilities provide an extra layer of protection for important and irreplaceable scientific data.

The "active archive infrastructure" technologies used by Fermilab leverage the unique wide-area transfer protocols and cached storage systems developed by the high energy physics community. These protocols are designed to move lots of data quickly, efficiently and without error between sites on the global network, and are integrated with Fermilab's high capacity hierarchical storage archive service. The archive facility at Fermilab is capable of providing access over a 100 Gb/s network to hundreds of petabytes of data, with long term persistence and custodial care, and with high performance retrieval.

The Fermilab Archival Storage Service provides a high level of data integrity. Several transfer protocols that support end-to-end checksums are available, the storage system monitors data integrity with automated random sampling and checksums are verified on every read. The data is also protected by policies and procedures against accidental harm, such as physically write locking the cartridges when they are full. In addition, files can be replicated to tape at a second data center on the Fermilab site for added protection. If you are interested in utilizing the Fermilab Archival Storage Service, please use the form at the end of this page.

Services Description

The Active Archive Facility is an Hierarchical Storage System consisting of Tape storage, called Enstore, with a front-end disk cache called dCache. A comprehensive description of the service and recommended use is available at Services description

Usage Information and Cost

Customers enter into a Strategic Partnership Project agreement with Fermilab. This agreement consists of a set of Terms and Conditions, and a Statement of Work. The Statement of Work specifies the responsibilities of both parties, and a cost model for 1 or more years. Yearly Storage charges are based on several quantities:

  • The total accumulated amount of tape storage used at the end of the year
  • The amount of tape media that needs to be acquired for the year
  • The amount of tape-drive hours used

At the beginning of each billing year, the customer supplies information on the estimated storage and drive utilization they expect to use, and are billed in advance for the year. If funds run out during the year, more must be supplied to be able to store additional data. If there are leftover funds at the end of the year, they are carried over into the next year. The portion of the yearly cost for media is spent at the beginning of the year, and the costs associated with the total amount of data stored on tape and the drive-hours are drawn on a monthly basis. The metrics that these charges are based on are available on the Customer's Active Archive web pages. These are updated daily, and history of monthly summaries are available.

Contact Us

If you are interested in knowing more about the Active Archive Facility, fill and submit this form:

Name *:
Email address *:
Request *:

Fields marked with * are mandatory. The request will not go through if they are left blank. No URLs are allowed in the request.