Technology robs words of their intrinsic meaning: what happened to “like” and “friend”? Is it true that Facebook patented “face”? What is “oracle”? Who is Cassandra — a NoSQL database or a Trojan princess, who had a gift of prophecy and whom Apollo cursed never to be believed?
At this week’s Cassandra Summit, I learned that engineers should think of github as their resume. And I found a new distance measure: buildings a millisecond apart. The 2012 Cassandra Summit attracted 830 attendees compared to 140 people just two years ago — think of this growth chart!
“I learned a lot,” — was a leitmotif of every speaker. I learned a lot too. This blog post is a brief recap of my most memorable takeaways (I am only reporting). For more on the Summit, see #cassandra12 and especially @Merv (please consider this a pseudo-tweet). (Abundance of parentheses in my writing is due to the years of Lisp in my past (hello, world of key-value pairs! (that is Cassandra (non-Greek.))))
Talking about Greeks, I’ll start in the Olympic style: a regular hard disk drive vs. a solid state drive (SSD) scored 27 to 4 milliseconds and $1.80 to $3.10 per hour on Amazon Web Services (AWS).
Amazon elastic cloud is out of stock on SSDs — Netflix took them all (did you know that Netflix is a hardware and PaaS company, and Amazon is one more Greek?) Capacity planning was another repeating theme. A rule of thumb, for example: always keep your disks half-full for a standard Cassandra cluster. With a rapid user population growth, performance gets spiky!
Cassandra data modeling session was standing room only and involved intricate column families (think Doric, Ionic or Corinthian columns). In-depth query tuning discussions were everywhere. Who said no DBAs necessary and data modeling is not an issue for noSQL?
Cassandra Summit was worthwhile attending even for a single session by @arunxjacob from Disney. I applauded his forward thinking about a common data platform. Netflix talked about a data platform too. PaaS for DaaS (or whatever acronym, as long as it’s a data platform) will soon become a big subject for corporations and for vendors.
The funniest Q&A of the day to my quirky sense of humor was:
Q: “How do you switch to new technologies?”
A: “You can do it unless you have a CTO or CIO with amazing vision.”
@Merv told me in this regard that sometimes he wants to raise his hand and say: “I am from Gartner, let me take this question.” So, I am from Gartner, and let me answer: Gartner does not recommend guerilla warfare as an IT strategy. New technologies are good when they are deployed for a reason not for the sake of technology. By deploying noSQL for a proper purpose with clear business outcomes, you will do a favor not only to your company but to Datastax too: there are enough use cases where Cassandra shines, and using it for everything just because you can will compromise this technology.
Disney, eBay and some other speakers told during the Summit that they bring Cassandra into play where it fits nicely: real-time analytics of social signals (what the most “liked” items are right now), time-series for fraud detection, log analysis to find common issues, multi-data center support and load isolation when you can set up a virtual data center to separate tasks (such as OLAP and OLTP). Cassandra is good for graphs. eBay creates taste graphs by items and users, 14+ billion edges. Titan, a new graph database, is one more Greek charmed by Cassandra. Most Trojans considered Cassandra insane — indeed, it is insanely fast.
And yes, I have seen big data fashion again! Cassandra t-shirts evaporated within the first hour of the Summit. And these are two new big data haute couture photos.
|“In Soviet Russia, cloud deploys you!” – could not pass this t-shirt.|
C* stands for Cassandra. The daughter of the Trojan king Priam turned into a noSQL database, and lost her name too. Is it because of the Apollo’s curse or because Greek gods meant well?
Read Complimentary Relevant Research
Organizing for Big Data Through Better Process and Governance
With big data past the Peak of Inflated Expectations on the Hype Cycle, organizations are addressing next-level challenges and asking,...
View Relevant Webinars
Hadoop and Spark: Understanding Open Source Opportunities and Risks
As companies build foundational data and analytics infrastructure with Spark and Hadoop, the market continues to shift and evolve in...
Category: analytics big-data cassandra catalyst catalyst-na cloud data data-paprazzi events fashion geospatial information-everywhere innovation inquire-within local-news market-analysis nosql open-source skills trojans
Tags: analysis analytics aws big-data big-data-adoption california catalyst data data-janitor data-paprazzi data-spy fashion haute-couture innovation market open-source pseudo-tweets search silicon-valley ssd vendors
Comments or opinions expressed on this blog are those of the individual contributors only, and do not necessarily represent the views of Gartner, Inc. or its management. Readers may copy and redistribute blog postings on other blogs, or otherwise for private, non-commercial or journalistic purposes, with attribution to Gartner. This content may not be used for any other purposes in any other formats or media. The content on this blog is provided on an "as-is" basis. Gartner shall not be liable for any damages whatsoever arising out of the content or use of this blog.