Google I/O: Android Interface, Cloud Advances Star

Day one of Google I/O lacked skydivers but delivered new APIs, tools, and connectivity to push Google's mobile device and cloud capabilities forward.

of a restaurant where a reservation had just been made. The handoff occurred without opening a new browser window.

Another use for Wear is to equip the watch with a Bluetooth signal that tells your phone you are present, which obviates the need to enter a PIN number when the phone's owner picks it up to use. If the owner were separated from the signal, the phone would lock itself against outside use.

Urs Hoelzle, senior VP of engineering and chief cloud architect, said Google has a new big data service that makes somewhat obsolete its MapReduce system, which has found popular use at such places as Amazon Web Services' Elastic MapReduce. In its place, Hoelzle recommends using Cloud Dataflow, a big data system capable of handling large streams of data or events in parallel pipes. It's designed to be easy to set up the management of complex data flows, since it can deal with both batch jobs and streaming data. At Amazon, MapReduce or Hadoop would handle the batch job, while Amazon Kinesis, introduced last November, would handle streaming data.

Hoelzle called Google cloud solutions architect Eric Schmidt 2nd (not Eric Schmidt, Google's executive chairman) to the stage as the Dataflow demonstrator. Schmidt used it to process data streaming out of a World Cup soccer match, looking for athlete performance anomalies, then alerting and reporting them to Dataflow managers.

Hoelzle also introduced four new developer tools for use in building applications to run on the Google Compute Engine or App Engine.

-- Cloud Save is a simple API for saving user preferences or application use information drawn from each user visit to Google Cloud Data Store, its storage service for unstructured data. Once there, it can be analyzed by Big Query or other analytics engine and the results used in connection with the operation of the application.

-- Cloud Debugging lets a developer who sees something malfunctioning in the application to inspect source code, insert a watch point when a suspected trouble area is executed, and then highlights the suspect code when the suspect activity occurs again. Greg DeMichellie, director of Cloud Platform product management, illustrated stray emoticons being inserted in the display of his WalkShare application, then discovered a regular expression error in the source. With the application running, the correction was inserted into a new build and incorporated into the production system.

-- Cloud Tracing is a performance oriented tool that allows a developer to inspect requests to an application and see which ones are running up clock time, affecting performance. A request taking 200 milliseconds to get a response is shown to be caused by a code loop, which DeMichellie corrected on stage. He then used Cloud Tracing to check the before and after latency of the application to make sure the app as a whole was running faster. It was.

-- Cloud Monitoring is a tool for gaining visibility into the operation of CPU, disk, and memory of a workload on Compute or App Engine, as most monitoring systems do. In addition, the tool can also monitor a running service and respond with alerts if service levels decrease below certain thresholds. The tool, for example, can monitor a running Reddis system and detect when its exceeded its memory threshold for longer than an allowed period.

Senior VP for Chrome and Android, Sundar Pichai, recently described by BusinessWeek as "the most powerful man in mobile," cited stats on how far Android has advanced over the course of the last year. Twenty-eight percent of the access to YouTube was from Android devices in 2013; 42% so far in 2014. Android's share of the smartphone market was 39% in 2012; 46% in 2013; and 62% so far in 2014, he said.

Handset manufacturers Micromax, Karbonn, and Spice have been working with Google to offer Android smartphones in India for under $100. Pichai said he has been using a Micromax model and "it's really good."

A protester interrupted Urs Holzle as he spoke, shouting, "You work for a company that builds machines that kill people," a reference most attendees appeared to be dumbfounded by. He was escorted from the Moscone West by security. Seconds later, the next speaker, Michellie, began his talk on WalkShare: "This is the most peaceful application that I know how to build," and got an appreciative round of applause.

InformationWeek's new Must Reads is a compendium of our best recent coverage of the Internet of Things. Find out the way in which an aging workforce will drive progress on the Internet of Things, why the IoT isn't as scary as some folks seem to think, how connected machines will change the supply chain, and more. (Free registration required.)

Editor's Choice
Samuel Greengard, Contributing Reporter
Cynthia Harvey, Freelance Journalist, InformationWeek
Carrie Pallardy, Contributing Reporter
John Edwards, Technology Journalist & Author
Astrid Gobardhan, Data Privacy Officer, VFS Global
Sara Peters, Editor-in-Chief, InformationWeek / Network Computing