PLANNING AND LEADERSHIP FOR A BETTER WORLD
PLANNING AND LEADERSHIP FOR A BETTER WORLD A new report from Aberdeen Research offers up more than a few interesting nuggets on Big Data, the challenges in processing and analyzing today's growing data loads, and how in-memory computing can play a critical role in speeding up the collecting, managing and sharing of data in the enterprise. At least for those enterprises than can afford its cost.
First, let me share some interesting statistics provided in the report:
● Business data is growing at a 36 percent clip every year.
● The top challenge cited in dealing with Big Data is the ability to get data faster (this is actually from a report the firm did in December 2011).
● Of 196 clients Aberdeen talked with about Big Data, 33 are using in-memory computing. The reason the majority aren't is likely cost.
● It takes about 42 seconds for a data query response with in-memory computing compared with more than 75 minutes at an enterprise not using the technology.
● In-memory can process 1,200 terabytes of data in an hour, compared with 3.2 terabytes an hour for enterprises that are not using in-memory computing. That's an efficiency gain of 375 percent.
In-memory computing, in the simplest of terms, makes processing, collecting and analyzing data fast, and that's a good thing for users, IT organizations dealing with increasing data growth and the business as data is the prime factor in business strategy decision making.
But like any technology, in-memory computing has its own set of unique quirks, issues, pitfalls and challenges.
First of all, it's not cheap. It needs high-powered servers, multicore processors and tons of RAM to work. It also needs appropriate software and analytical applications. All those pieces are needed for the processing speed as the technology allows for terabytes of data to be stored directly on a server's RAM, and not a stored disk somewhere, which ensures no latency.
While it's hard to get vendor pricing on in-memory appliance costs, and the report doesn't provide any pricing data, just consider this report statistic: Enterprises using in-memory computing spent about $850,000 on storage and data processing hardware in the last 12 months.
Another issue with in-memory computing is that it doesn't embrace all data. It's more specific to structured data than unstructured and does its best work on data sets in transactions, sales figures, customer information and things like product codes.
If your company has the money and realizes the inherent value of data in today's business strategy, in-memory technology might be your best option.
IBM expands global cloud capabilities with advanced #SmartCloud services & new customer adoption http://t.co/qpbR1XxK #ibmcloudStill confused to think of your #cloudmoment! Here are some good quotes that can help you think of your own http://t.co/3dsI0nO6 #ibmcloudRT @Shubhs77: I experience #cloudmoment whenever I upload my Daddy's fav songs/files on the Dropbox installed on his phone. He just need ...Good morning @cloud_indonesia! When have #cloud touched your life? Share your #cloudmoment! Check out our more at http://t.co/J4nxZKQZWe've launched a new site with stories of the moment cloud "clicked" for folks. Check it out & share your #cloudmoment! http://t.co/J4nxZKQZGood Morning from Asia/Pacific! #ibmcloudThat's it for me! More updates from the #IBMcloud Forum Chicago coming your way tomorrow. Over to Asia/Pacific.@RCChicago Thank you!*Brand new* video: Innovate without boundaries and harness the power of cloud >> http://t.co/gxSPLFbL #ibmcloudLots of #cloudmoment stories here today @RCChicago. http://t.co/X1JRs370
Like us on Facebook 
Follow us on Twitter 
Get the Smarter Tech Newsletter 
Register to view the archived "Data Center Optimization Techniques: Stop Wasting Time & Money" eSeminar
Register Now > 
