SiSense Unveils Crowd Accelerated Analytics

The typical scaling problem that businesses see as their data operations grow is that the more people there are using the system, the more sluggish the system becomes. Israeli startup SiSense says it’s taken this challenge head-on, creating a new paradigm the call “Crowd Accelerated Analytics,” which it says provides faster query results as loads increase.

“Until now, the conversation has been around scaling data,” said Eldad Farkash, Co-Founder and CTO of SiSense in a statement. “But what about scaling massive amounts of users and queries? Old school BI nets users limited answers to limited questions. Crowd Accelerated Analytics lets you analyze billions of rows of data in a flash.”

According to SiSense, the system works by learning from results of similar (but not identical) queries that are being made in the system. Forbes’ Gil Press got to the bottom of the technology in a recent article, explaining how it worked:

“With SiSense technology, all queries are broken down to more granular, machine-level “instructions.” This detailed knowledge of the nature of each query helps it understand which data to keep on the disk and which data to bring to memory, at times even anticipating the next query before the user asks the question. It then uses sophisticated mathematical calculations to process the query in parallel on the CPU.”

According to Press, this ability to translate each query into machine-level instructions helps in accommodating larger numbers of users. “Instead of looking for identical queries like other caching systems we look for similar queries with 80% overlap in the instructions,” SiSense CEO Amit Bendov told Press. “Every query is broken down into a very large tree structure and we look if this tree has sub-trees that are identical or similar to other queries. It’s actually a learning system that stores all the answers to the most difficult question.”

“While other solutions get clunky and costlier as load increases, SiSense Prism gets faster and overcomes the increased expenses associated with slow-downs,” said Bendov in a statement. “It’s like we’re breaking the laws of physics. We call it power querying.”

The company last week released version 4.7 of its “In-Chip” Prism 10x analytics platform, which the company boasts provides non-technical users with the ability to analyze 100x more data at ten times more speed than current in-memory offerings. In order to deliver on such claims, the developers have constructed the query kernel to run in the L1 cache, enabling caching algorithms to decide in real-time how to use machine capacity to store, compress, and access the data.

Improvements with its 4.7 release include new IBM DB2, Teradata, and Salesforce.com connectors as well as some nice visualization widgets that give users expanded options in visualizing data.

SiSense has garnered quite a bit of attention for itself in the last 12 months, after having spent several years developing the ElastiCube technology which sits at the heart of its analytical package. Founded in 2004, the company spent roughly five years working on the product before it was ready for release. In February of this year, it was selected as the Audience Choice Award recipient at the O’Reilly Strata conference “Startup Showcase.”  In April the company took in a $10 million dollar venture funding round led by Battery Ventures.

Company officials say that SiSense is currently in a period of growth, having grown 300% from two years ago, and 520% in 2012.

Related items:

Is “In-Chip” Data Processing the Next Revolution? 

SiSense Wins Audience Choice Award at Strata 

Five Interesting Startups at Strata