"Do you want to become a farmer?!” In a sense, yes.
Two years ago an informal group met for drinks in downtown Palo Alto: a mix of grad students, investors, and data science experts in Silicon Valley. In the back and forth of our conversation, we took turns describing planned projects. At the time, prominent VC firms were racing headlong into health care ventures. Much of our group seemed pointed in that direction.
In my turn, I mentioned one word: Agriculture.
That drew laughter, “You want to become a farmer?!”
In a sense, yes.
Impact of data science beyond silicon valley
Practices involving large-scale data, machine learning, cluster computing, etc., toppled entire sectors over the past decade. Retail (Amazon) went first, followed closely by Advertising (Google). Automotive (Tesla) may be next. Clearly, the impact of data science has moved beyond Silicon Valley, with mainstream industries leveraging data that matters… not simply to improve marketing funnels, rather to overhaul their supply chains, manufacturing, global deployments, etc. Advances in remote sensing and “Industrial Internet” accelerate that process, with IoT data rates growing orders of magnitude beyond what social networks have experienced, compelling new technologies.
Sometimes when a group of insiders starts guffawing, there is perhaps a subtle point being missed. Consider that Silicon Valley has spent the past decade extracting billions from e-commerce, ad-tech, social networks, anti-fraud, etc. Extracting is the quintessential word there. I wondered: among the industries outside of Silicon Valley undergoing disruptions due to large-scale data, where did Agriculture fit? Why did it seem laughable to experts as a data science opportunity?
In the summer of 2012, Accel Partners hosted an invitation-only Big Data conference at Stanford. Ping Li stood near the exit with a checkbook, ready to invest $1MM in pitches for real-time analytics on clusters. However, real-time means many different things. For MetaScale working on the Sears turnaround, real-time means shrinking a 6 hour window on a mainframe to 6 minutes on Hadoop. For a hedge fund, real-time means compiling Python to run on GPUs where milliseconds matter, or running on FPGA hardware for microsecond response.
With much emphasis on Hadoop circa 2012, one might think that no other clusters existed. Nothing could be further from the truth: Memcached, Ruby on Rails, Cassandra, Anaconda, Redis, Node.js, etc. – all in large-scale production use for mission critical apps, much closer to revenue than the batch jobs. Google emphasizes a related point in their Omega paper: scheduling batch jobs is not difficult, while scheduling services on a cluster is a hard problem, and that translates to lots of money.
Mesos offers reliability, efficiency and faster developer productivity.
Virtual machines (VMs) have enjoyed a long history, from IBM’s CP–40 in the late 1960s on through the rise of VMware in the late 1990s. Widespread VM use nearly became synonymous with “cloud computing” by the late 2000s: public clouds, private clouds, hybrid clouds, etc. One firm, however, bucked the trend: Google.
Google’s datacenter computing leverages isolation in lieu of VMs. Public disclosure is limited, but the Omega paper from EuroSys 2013 provides a good overview. See also two YouTube videos: John Wilkes in 2011 GAFS Omega and Jeff Dean in Taming Latency Variability… For the business case, see an earlier Data blog post, that discusses how multi-tenancy and efficient utilization translates into improved ROI.
One takeaway is Google’s analysis of cluster traces from large Internet firms: while ~80% of the jobs are batch, ~80% of the resources get used for services. Another takeaway is Google’s categorization of cluster scheduling technology: monolithic versus two-level versus shared state. The first category characterizes Borg, which Google has used for several years. The third characterizes their R&D goals, a newer system called Omega.