RainStor raises $12M to turn your big data small
#SuryaRay #Surya RainStor, a database vendor that focuses on extreme compression of large data sets, has raised a $12 million Series C round from Credit Suisse and Rogers Venture Partners as well as existing investors Doughty Hanson Technology Ventures, Storm Ventures and The Dow Chemical Company. RainStor is riding quite a wave of momentum right now, no doubt thanks to claims it can reduce data volumes by at least 95 percent using its unique compression and de-duplication technology.
The company focuses on historical data that might need to be stored for long periods of time and isn’t likely to change. In some cases that might be data stored for regulatory compliance, while in others it might be machine data such as server logs that would never be written over in the first place. RainStor has also beating the drum pretty loudly around big data, where it certainly has a compelling proposition.
Because RainStor utilizes massively parallel processing, it can ingest and query data in a hurry. The company claims ingest speeds of 30,000 to 50,000 records per second per core. Query results also return faster because the analyses are taking place over such a smaller volume of files.
RainStor also can sit atop Hadoop for users that want to churn through unstructured data via MapReduce or Pig jobs as well as run more-traditional SQL queries. This type of hybrid system is becoming a hot topic in big data circles, and is the premise of a handful of other products, including Boston-based startup Hadapt. I’ll actually be speaking about the intersection of Hadoop and SQL at our Structure: Europe conference on Oct. 16 with Cloudera’s Amr Awadallah and NuoDB’s Barry Morris.
RainStor says it can provide 50 to 80 percent smaller Hadoop clusters and increase performance by up to 100 times. These are both big concerns as data volumes continue to explode. Quantcast created its newly open sourced distributed file system so the company could bring operational expenses under control while also lowering the time it takes to process its ever-expanding data set.