Here’s the problem. As Fusion-io sees it, virtualization is not seeing its potential. There’s a bottleneck. Why? Data is getting stopped up due to a lack of processing capability. To solve the problem, companies are developing complex storage systems, using intricate disc network to alleviate the memory issue. The result is a storage environment that continually needs more hardware, more servers to solve problems with performance.
Fusion-io uses a combination of hardware and software to create a storage memory platform that moves the data closer to the CPU where it is processed. It’s fusing the memory and the storage into a virtual storage layer (VSL).
IO Turbine offers a complimentary solution. It takes advantage of the flash local to the server. It also holds the data close to where it is used.
This means that the CIO can take a second look at those expensive, performance-oriented network storage systems and memory appliances that struggle to deliver performance from the far end of a network. Instead, a company can use a shared storage system that takes advantage of its combined strength to get better performance.
This shared storage environment also means that it is feasible to virtualize data intensive applications like ERP systems without sacrificing performance. Could a company use the Fusion-io and IO Turbine technology to virtualize all of its applications? Arguably, the answer is yes.
“We do for data what VMware does for virtualization,” said Fusion-io Chairman and CEO David Flynn in the earnings call yesterday.
Flynn said in the earnings call that there is about $260 billion in wasted spending on server sprawl due to this problem of data bottlenecks.
What results is a data supply issue that just holds everything up.
“This data supply problem is ilk a factory of workers sitting on their thumbs waiting for supplies to arrive,” Flynn said.
Its this capability to process data faster that Fusion-IO can call Apple and Facebook strategic customers.
Fusion-io discussed Apple a strategic partner, alongside Facebook. Apple and Facebook both emphasize the importance of user experience and rapid response time. Both are using Fusion-io technology to help achieve that by putting the key databases in persistent memory as close to the server as possible and are using Fusion-io’s VSL technology to minimize I/O overhead. This architecture is discussed in detail in a previous Wikibon post entitled “Little’s Law and Lean Computing.”
CIOs and CTOs need to take a look at how they can virtualize the entire enterprise. The need is there for better performance. The services organizations that can tell this story well will be the ones best positioned for driving customers to the next generation of virtualization technology.