• Reliable: SAFS provides automated fail-over processing to ensure quicker and more reliable file transfers than previously possible.
  
  
• Fast: Because interactions with the firewall and security protocols are reduced, data is delivered in near real time making optimal use of bandwidth.
  
  
• Unattended operation: This automated system includes self-correction capabilities to enable operation with no human interaction.
  
  
• Scalable: Centralized data storage is easily scalable by adding more storage capacity.
  
  
• Universal: The SAFS code is generic and easily adaptable to most hardware and operations requirements.
  
  
• Reduces costs: Automated Web reporting eliminates the need for expensive dedicated telemetry equipment and network connections. Additional cost savings are found in unattended operation and the use of COTS hardware and software.

• File server systems and operating systems
• Digital movie data distribution (i.e., delivery method by streaming data over broadband networks from centralized servers)
• Commercial satellite downlink distribution
• Financial market distribution
• Traffic management centers
• Air traffic control centers
• Medical information distribution
  
  

The Standard Autonomous File Server performs automated management of large data files and provides quick access to these files without interfering with the assets involved in their acquisition.  Assets are the systems or components that acquire the data and parse the information into files. The SAFS infrastructure separates the data acquisition activity from the file distribution activity.

The system uses a fail-over data file transfer process that eliminates unnecessary network congestion when transfers are unsuccessful. With the use of the COTS software, FASTCopy, SAFS’ provides:

• Self-correcting transmissions for file delivery
• Recovery from point of failure
• Stop/resume transmission control
• On-the-fly compression
• Programmable network bandwidth control

How it works

SAFS servers are located on both open and closed networks. The closed network SAFS server acquires the data and provides a backup location, while keeping it isolated from clients. Data is transferred to a second SAFS server located on an open network with the clients to provide a single contact point for project data, as well as to eliminate special firewall and security rules and speed up transmission.

Clients can either “pull” files or receive “pushed” files from the SAFS system. When a file transfer is successful, the client sends back a receipt confirmation notification (RCN) with a success or error status. To increase the chances for successful transfers and to reduce network congestion, SAFS’ fail-over file transfer process enables the client to specify multiple backup receiving locations and a specific number of attempts to be made at each location before a failure to deliver notice (FDN) is sent. The SAFS system first attempts to transfer the file to the client’s primary receiving location. If unsuccessful after the specified number of tries, it self-corrects and moves to the first backup location and again attempts to transfer the file the specified number of times. This process is repeated at any remaining backup locations until the transfer is successful or there are no more backup locations to attempt.  

If a file transfer is ultimately unsuccessful, an FDN e-mail is sent to the client so the client can pull the file from the sender’s location during a specified retention period.

Prioritization

Because multiple projects (and sometimes single projects) have a high likelihood of simultaneous file transfers and bandwidth issues, a prioritization system was developed to help eliminate this problem. Rather than using a first in first out (FIFO) scheme, the SAFS system combines bandwidth control with four levels of file prioritization. Those are: special (for emergency information), high (for files requiring fast delivery), medium (for files with normal delivery needs), and low (for the least urgent files). There can also be subcategory priority levels for further classification of documents and projects.

Before a file is transferred from the SAFS system, its priority is compared against that of any file already being transferred from the system.  If the new file’s priority is higher, then the transfer of the lower priority file is stopped, and the higher priority file is transferred.  When the higher priority file transfer is complete, the previously stopped transfer is resumed from the point at which it had been stopped rather than from the beginning of the file. If the new file’s priority is equal to that of a file being transferred, then the new file can either be queued until the previous file’s transmission has completed, or the files can be transferred together by sharing the bandwidth equally.

Performance reporting

SAFS performance reporting eliminates the need for expensive dedicated telemetry equipment and network connections. Information from each RCN is added to the SAFS event log for each file and is used in performance monitoring and latency reports. These status reports are available on each SAFS system and are sent to a Web server for easy access. These reports enable clients, project managers, and system administrators to remotely verify file availability and system performance.

Why it is better

Prior to the development of the SAFS technology, clients requested data directly from the systems acquiring the data, which were located on closed networks and, therefore, necessitated special firewall or security exceptions. This caused excessive use of the operations bandwidth when multiple clients simultaneously made requests for the same data. In addition, the file transfer protocol used by operations could not recover from network problems or customer disk errors. A SAFS server is located outside of those closed networks for data distribution to clients, so the transmissions are faster and necessary bandwidth is reduced.

SAFS’ automation reduces costs and personnel, as well as adds flexibility for clients. Because the system incorporates self-correcting transfer methods, there is a time/cost savings realized when transfer problems occur, which would require a live person’s action on non-automated systems. Personnel and cost savings are also realized at the client level, because data pushed or pulled from SAFS also provides automation opportunities there. By using redundant systems, there is additional reduction in costly down time.

The SAFS system was developed to reduce latent delivery time for satellite operations and finds uses in many other commercial applications. Since its initial development, it has been in use in six locations supporting eight satellite projects with high success rates. SAFS offers flexible, reliable, and timely data distribution at lower costs.

System configuration

Hardware can be configured in a standard 19” rack and includes the following:

• RAID storage system
• RAID monitoring/configuration system
• Server
• Rack mounted keyboard/monitor/touch pad (optional)

The SAFS system also contains the following software components:

* The web reporting feature is configurable.

This technology is part of NASA’s Innovative Partnerships Program Office, which seeks to transfer technology into and out of NASA to benefit the space program and U.S. industry. NASA invites companies to consider licensing the Standard Autonomous File Server (GSC-14409-1) technology for commercial applications.

For information and forms related to the technology licensing and partnering process, please visit the Licensing and Partnering page. (Link opens new browser window)

If you are interested in more information or want to pursue transfer of this technology (GSC-14409-1), please contact:

Innovative Partnerships Program Office
NASA Goddard Space Flight Center
E-mail: SAFS@gsfc.nasa.gov