There is a lot of hype – and confusion – about the growing phenomenon of ‘Big Data’, a.k.a. the acquisition, analysis and interpretation of ridiculously huge data sets. Depending upon the sources you use, Big Data is still very much in its infancy, but it is growing and will have a major impact on network traffic, and on WAN optimization, in the near future.
According to a recent Deloitte survey, the 2012 Big Data market will be worth between $1.3 and $1.5 billion, primarily consisting of pilot projects with fewer than 50 full-scale Big Data projects (10 PBs and above) worldwide. Another survey, from Wikibon, puts this year’s Big Data market at just over $5 billion in total factory revenue, and growing to $50 billion by 2017. Deloitte predicts that more than 90% of Fortune 500 companies will have a Big Data initiative under way by the end of 2012.
It gets better – or at least more ‘interesting’ – from a network perspective. Big Data is expected to account for more than half of the world’s data in the next five years, according to a study from Internet Research Group. The study suggests this growth has the potential to clog networks, with the bandwidth demand compromising the latency, speed and reliability of the enterprise WAN.
“If it needs to traverse the WAN, it [big data] needs to be factored into QoS schemes, WAN optimizer prioritizations, and possible storage array replication schedules to make sure it gets where it needs to go when it needs to go there with as little impact as possible on other critical WAN traffic,” said John Burke, Principal Research Analyst at Nemertes Research.
That’s where a good WAN optimization solution can play a big role. According to Silver Peak, the five areas where WANop can help keep networks running quickly and reliably are:
- Throughput – WAN capacity that can simultaneously handle data reduction, compression, QoS, and latency/loss mitigation at high WAN speeds (up to multiple Gbps in a single appliance);
- Storage capacity –disk-based WAN deduplication will provide superior memory retention and performance;
- High flow counts – WAN optimizers that are not limited by the number of transport-layer (TCP, UDP, RTP) flows in a network;
- Wirespeed encryption – WAN optimizers that leverage state-of-the-art hardware will ensure that encryption does not adversely impact scalability and performance; and,
- Application breadth – WAN acceleration that operates at the network layer of the ISO stack will provide data reduction, compression, and other techniques across all enterprise applications, including SQL, VoIP, and video.
With network performance as much as 90X better with WAN optimization, it looks like the expected Big Data surge will spur additional WANop use, too.
image source: flickr (Steven Vance)