There are two inexorable facts of life when it comes to data. The first is that the amount of data that needs to be managed is growing faster than most IT organizations can cope with. The second is that as the size of the files containing that data continues to grow, the performance of our systems and networks starts to slow.
This latter issue is becoming especially acute with the advent of next-generation virtualization, under which entire applications can move around the network only to discover that the data they need to access that was once local now needs to be transferred half way around the world.
This concept of the virtual data center requires a whole new approach to transferring data, which is exactly what the folks at Aspera are trying to drive in the form of a new high-speed protocol for transferring data.
Although primarily used as an alternative to FTP running over a TCP/IP network because Aspera’s Faspex software is about 1,000 times faster, interest in Aspera’s technology is growing in the storage community with companies such as EMC and IBM adding support for it into their offerings.
According to Aspera co-founder and president Michelle Munson, the days when data transfer rates become a major problem in the enterprise are close at hand. But rather than throw a whole lot of expensive new infrastructure at the problem, Munson says IT organizations would be better off examining the efficiency of the technologies being used to transfer that data today.
Of course, vendors always prefer to sell new expensive hardware. But when push comes to shove, there may not be any problem with the hardware at all. It’s just that the software running on it is so cumbersome to process. It requires massive amounts of horsepower, not to mention energy, to run.
.Naturally, that begs the question, how often is this the case? But rather than take on the entire software development community at once, maybe we can start with a little fundamental improvement to the way we transfer data. Who knows, that alone might save enough money to pay for upgrades to more efficient applications elsewhere.
