While many companies that support the space promote impressive file transfer speeds as the main attribute of a successful multi-platform production environment, Michelle Munson, president and co-founder (with husband Serban Simu) of Aspera, thinks that’s only a small part of the puzzle. With dual B.Sc. degrees in Electrical Engineering and Physics from Kansas State University, and later excelling as a software engineer in research and start-up companies including the IBM Almaden Research Center before founding Aspera in 2004, Munson has devoted most of her career to file transfer optimization and network bandwidth management.
A self-termed “technologist,” she said technology considerations should never be the first element of concern for a production team leader or network system designer. What she and her Emeryville, CA-based company sees is a need to clearly understand the workflow at hand and then conform network protocols to fit a specific business model.
“Speed is not the only thing,” she said. “You have to look at your business goals and how that content file will be used and leveraged for multiple uses. Then you have to think about what you'd want to have happen with that file, given that it's the primary asset, and then from that you figure out the technical requirements. But technology considerations are never the first element of concern.”
Security and the ability to accurately track a file are also key, Munson said, as are unlimited, instant access to media, security, bandwidth management, reporting, bandwidth control and high-performance collaboration. Basically, production teams expect certain things to happen, and they must.
“Reliability is also critical,” Munson said. “We like to compare our products to the familiar overnight package delivery model. When you send a package, you have an expectation of when the package will arrive at its destination. Likewise, the person receiving that package expects it to be there when they need it. The same holds true for successful electronic file transfers and content sharing.”
Customers often compare the services that Aspera and others offer to a typical Internet-based file transfer service like Drop Box, but they are seriously mistaken, Munson said.
“Internet services are not adequate,” she said. “Sending large files over long distances has always been a problem. Then there’s the issue of sending multiple files simultaneously, which must be synchronized and delivered in a highly reliable way. Services like Drop Box can't do that effectively. All of the Internet services are built on top of TCP (like FTP and HTTP, developed in the 1970s and '80s). We started Aspera when we recognized that TCP is more or less unusable as a file or bulk date transfer mechanism. It has very severe bandwidth under utilization due to its flow control properties. Effectively, if you tried to use it across a long distance, transfer speeds are very slow. That means large data sets are impractical for sharing and collaboration among a production team.”
What Aspera recommends is finding a happy medium between an on-premise software on a standard client computer, and hosted, cloud-based services like Amazon. Many of its customers —like Encompass, NetFlix and several major European sports leagues — that have decided to use a cloud environment have made that decision based on available resources as well as how much scale and processing power they anticipate needing. The cloud allows them to expand and contract quickly, as the need arises, and support individual business models.
To help the process, Aspera — more accurately Munson and her husband — has developed its Fast Adaptive Secure Protocol “fasp” file transfer technology, which it builds into many of its software products, including the Enterprise Server, Connect Server, Point-to-Point and Client 3.3, faspex 3.5 and Shares 1.5, Console 2.0 and Aspera Sync 1.4, as well as a new Faspex Mobile Client for Android (all of which will be exhibited at the upcoming IBC show in Amsterdam). There’s also an upgrade available to the Aspera SDK on the Aspera Developer Network. Fasp features all of the properties of TCP, but is optimized to move large amounts of data over IP networks with “predictable throughputs and delivery times.”
Munson said the patented transfer software eliminates the fundamental shortcomings of conventional TCP-based file transfer technologies, facilitating fasp transfers that are hundreds of times faster than FTP/HTTP. The technology also provides a guaranteed delivery time regardless of file size, transfer distance or network conditions, including transfers over satellite, wireless, and inherently long distance and unreliable international links. And, the technology provides control over bandwidth utilization, transfer rates and bandwidth sharing with other network traffic. Security is built-in, including endpoint authentication, on-the-fly data encryption and integrity verification.
Of course, getting files to the right person at the right time quickly should not be underestimated. The company claims that over gigabit WANs with one-second RTT and 5 percent packet loss, fasp achieves 700Mb/s to 800Mb/s file transfers on high-end PCs with RAID-0 and 400Mb/s to 500Mb/s transfers on commodity PCs. Large data sets of small files are transferred with the same efficiency as large single files. The fasp software is very lightweight, and thus does not require specialized or powerful hardware in order to perform properly.
Some of the things that make it so fast, Munson said, are that special algorithms ensure that the effective data throughput does not drop due to network delay or when packet loss probability increases. It does this intuitively, without changing or adjusting data rates. It also guarantees good bandwidth sharing.
So, for Munson, what has changed over the past 10 years in the area of file transfer technology?
“The single most important change in this industry is knowledge,” she said. “When we started working with file transfer optimization, the fundamental technical problem was not understood. Furthermore, the opportunity to do this very highly performing network-based data transfer was not believed possible. Today, the industry is extremely well educated to the issue and developing solutions to address customer concerns.”
With the advent of 4K UHDTV, file sizes will continue to grow larger, adding additional strain on the network. But Munson again said that the real solution lies not with higher speeds or file sizes. It should be about the media organization’s workflow and business model. The networking technology should be flexible enough to fit the application, not the other way around.
“Everyone’s trying to digitize their time and resource-intensive processes,” Munson said. “What this is really all about is the digitization of what would otherwise be a physical process. We’re striving to attain in software and across networks what can be done in that easy and familiar overnight delivery scenario. As a user, I don't care how the file gets to me; it just better be there when I need it.”
