Developers using Google Cloud Platform can now talk to each other, and to Googlers, in a brand-new Google+ Community.



The new community is where we will share information about official Google Cloud Platform developer programs and events with you, to help more people get engaged. You can also get help from your peers, share your own stories and contributions, and find great developers to invite to join your project.Read More

Developers using Google Cloud Platform can now talk to each other, and to Googlers, in a brand-new Google+ Community.



The new community is where we will share information about official Google Cloud Platform developer programs and events with you, to help more people get engaged. You can also get help from your peers, share your own stories and contributions, and find great developers to invite to join your project.



Along with our other forums, such as StackOverflow, the Google+ community helps newcomers find the right resources more quickly.



How can you contribute? There are many ways:

• If you blog or know of any articles about Google Cloud Platform, you can share them with your fellow cloud developers in the new community.


• This also goes for podcasters and video bloggers (vloggers). If your content is relevant to the Google Cloud Platform community, you can share it here.


• If you are a public speaker, you can use the community to share a link to the event where you’ll be giving your relevant talk.


• If you spend a lot of time helping people with technical questions, you can use the community to help new folks find answers and get started, for example by pointing them to the correct StackOverflow tag.


• Got other kinds of relevant content? Share it here! Tutorials, podcast episodes, and videos of presentations about Google Cloud Platform—they’re all welcome in the new community space.

We’re just starting out with this new community, but we have big plans, and we want your ideas for making this space more useful and engaging for you, too. Share them with us in the community.



-Posted by Alex Maier, Google Cloud Platform Community Manager

Do you need an environment that seamlessly scales to meet demand? Do multiple teams develop your app using a variety of programming languages? Do you need to mix managed applications and self-serve virtual machines running a variety of separate services? Google Cloud Platform has you covered on all fronts.Read More

Do you need an environment that seamlessly scales to meet demand? Do multiple teams develop your app using a variety of programming languages? Do you need to mix managed applications and self-serve virtual machines running a variety of separate services? Google Cloud Platform has you covered on all fronts.



Take a look at our latest paper to learn how to overcome the challenges of deploying large applications in Cloud Platform. Our paper discusses a solution that shows how to use the right tools for:

• Different developer expertise


• Different data storage requirements


• Interactive and batch processing requirements


• Custom analysis engines


• Logically separate business components

Our paper walks through a real-world scenario of building a video sharing community. In this scenario, users upload videos from a mobile phone, the content is transcoded for playback in multiple formats and is presented on a web community site for watching and commenting.

Our sample solution showcases Google App Engine applications written in PHP, Java and Python. We use Google Cloud SQL for storing video metadata, to capitalize on the popularity and reduced learning curve of MySQL, and we store the videos in Google Cloud Storage for speed and scalability. The mobile client is written in Java and runs on Android. Hadoop clusters running on Google Compute Engine analyze and add sentiments to users' comments. And the whole shebang is orchestrated by another App Engine application. To learn how to build a solution like this yourself, get started now by reading all about it.



-Posted by Brian Lynch, Solutions Architect

We’re back with Google App Engine 1.8.5, and this release focuses on an upgrade to the Search API and improvements to the PHP runtime. Next week we’ll be at Zendcon ...Read More

We’re back with Google App Engine 1.8.5, and this release focuses on an upgrade to the Search API and improvements to the PHP runtime. Next week we’ll be at Zendcon, one of the largest conferences for PHP developers, where we’ll have some cool things to share about the PHP runtime.



Search API is now Generally Available

Surfacing Google technology to developers everywhere is one of our goals with Cloud Platform. The Search API does exactly this. With just a few lines of code, you can use Google Search technology to index and query millions of documents that contain structured data. Today, we’re making the Search API Generally Available so any developer can dive in and get started.



To understand how easy it is to get going with the API, let’s say you were a real estate company that wanted to add search capabilities. To do so, we’ll first import the Search library, and add some listings (referred to generically as Documents in the Search world).

from google.appengine.api import search

listingLocation = search.GeoPoint(37.78, -122.39)
listing = search.Document(
    fields=[
       search.TextField(name='description', value='Great condo in the city'),
       search.NumberField(name='bedrooms', value=2),
       search.GeoField(name='location', value=listingLocation)
       # omitting the other fields for this example 
       ])

Next we add the newly created listing to our index using the put() method.

try:
    index = search.Index(name='listingsForSale')
    index.put(listing)
except search.Error:
    logging.exception('Make sure you handle this error')

After we’ve added documents to an index, we’ll grab an index and specify our search query.

index = search.Index(name='listingsForSale')

# search for listings within 8050 meters (~5 miles) of the 94109 zip code
query_string = 'distance(location, geopoint(37.7929, -122.4212)) < 8050'

Finally we’ll perform the search and iterate over the results.

try:
    results = index.search(query_string)
    # Iterate over the documents in the results
    for scored_document in results:
        pass  # handle results
except search.Error:
    logging.exception('...and this one too')

And that’s just a small example. The Search API allows developers to:

• index Atom, Text, HTML, Number, Date, Geopoint fields


• perform queries across all field types, including geospatial queries, partial text matching


• score and sort results

We have many customers using the Search API for a variety of use cases. Key Ingredient, for example, uses the Search API to allow users to perform searches over the 1.6 million recipes in their database. Harlan Beverly, their CEO, said, “Our users are often shocked and delighted by the speed of our search results and our site overall.”



By moving to GA, the Search API is fully supported and comes with a monthly 99.95% SLA. For more information, be sure to check out our docs (Python and Java).



Improvements to the PHP runtime

The Cloud Platform team focuses a lot on providing great developer tools so that you can be as agile as possible in development. In this release, we’ve packaged a PHP interpreter binary in the Windows and Macintosh SDKs, including integration with the App Engine launcher, so that you no longer need to install PHP yourself.



We also believe modularized applications are the right way to go for architecting solutions in the cloud. So the PHP runtime now supports App Engine modules – the ability to split out large-scale applications into logical components that are able to share stateful services and communicate securely.



Finally, we’ve added the Logs API to the PHP runtime, which provides you access to the application and and request logs for your application.



That’s a lot of PHP goodness in 1.8.5, and we’re accepting more users, so sign up to give it a try.



Eclipse Tooling

We took your feedback that our Google Plugin for Eclipse should fully support Cloud Endpoints in the Eclipse standard Web Tools Platform and Java EAR files which are familiar to many Java developers as the most common pattern used in Eclipse for on-premise and cloud environments. Now, check out how to add an App Engine backend to your Android or iOS mobile app.



The complete list of features and bug fixes for 1.8.5 can be found in our release notes. For App Engine coding questions and answers check us out on Stack Overflow, and for general discussion and feedback, find us on our Google Group.



-Posted by Chris Ramsdale, Product Manager

Today's guest post comes from Dr. Chandra Krintz, co-founder and CTO of AppScale Systems. AppScale Systems is the team behind AppScale, the open source cloud platform that brings portability and failover to App Engine applications. ...Read More

Today's guest post comes from Dr. Chandra Krintz, co-founder and CTO of AppScale Systems. AppScale Systems is the team behind AppScale, the open source cloud platform that brings portability and failover to App Engine applications.



Recently, there have been interesting discussions around lock in and Google App Engine. With hosted cloud services, there is always a concern that the developers will find the barrier to exit too high. We have partnered with Google to relieve developers of the burden of porting their App Engine apps to other deployment targets via our cloud platform, AppScale.



App Engine developers can now execute their apps either on App Engine in the cloud using Google’s infrastructure or on AppScale over resources they control. To enable this, AppScale emulates App Engine and mirrors its APIs using proven and distributed open source technologies.



The video below shows how to setup AppScale and deploy your App Engine apps over Google Compute Engine. Users can also deploy AppScale over virtualized clusters, on-premise data centers, and public and private cloud infrastructures like Amazon EC2, Eucalyptus, and OpenStack. Deploying the platform consists of downloading or creating a virtual machine image with the AppScale code base, and using the AppScale tool chain to start the platform services across a number of virtual machine instances.







AppScale automatically deploys, configures, and scales your apps and their service ecosystems, so that you don't have to. Because the platform is open source, you can also customize and optimize the system according to your needs, and we can help.



AppScale Systems is a Google Cloud Platform Technology Partner and collaborates with Red Hat and Google on the Test Compatibility Kit (TCK) project. Learn more about AppScale and try it today at AppScale.com. Follow us on Twitter (@appscalecloud) or join our community of contributors and ask questions via the AppScale Community Google Group and IRC channel (#appscale).



-Contributed by Chandra Krintz, co-founder and CTO of AppScale Systems

Our developer community often creates interesting tools to customize their experience with Cloud Platform. The team at Streak built their Gmail-based CRM product using Google App Engine. Their founder, Aleem Mawani, blogged about how they created a system of audit logs to debug their usage of the ...Read More

Our developer community often creates interesting tools to customize their experience with Cloud Platform. The team at Streak built their Gmail-based CRM product using Google App Engine. Their founder, Aleem Mawani, blogged about how they created a system of audit logs to debug their usage of the App Engine Datastore.



Read Aleem’s post on the Streak developer blog to learn how they did it.



-Posted by Zafir Khan, Product Marketing Manager

One of the most common and yet difficult requirements that applications deal with today is sending email. Google Cloud Platform provides a variety of built-in solutions for sending mail like the ...Read More

One of the most common and yet difficult requirements that applications deal with today is sending email. Google Cloud Platform provides a variety of built-in solutions for sending mail like the mail.send_mail() API in App Engine. However, sometimes it’s easier to take advantage of a third party solution that puts 100% of its energy into solving that task. That’s where SendGrid comes in. We’ve been working closely with SendGrid to increase email deliverability and enable support for Google App Engine including testing, SDK improvements, and tutorials to help you get started.



Now, we are extending SendGrid support so that it also works with Google Compute Engine, our virtual machine based Infrastructure-as-a-Service (IaaS) offering.



SendGrid has put together some great content to get you up and running sending transactional emails from Compute Engine virtual machines. Learn even more on their blog. Also, get started with a great offer of 25,000 free emails a month* and let us know what you think.



*Google will be compensated for customers who sign up for a non-free account



- Posted by Brian Goldfarb, Head of Marketing

Google BigQuery is designed to make it easy to analyze large amounts of data quickly. This year we've seen great updates: big scale JOINs and GROUP BYs, unlimited result sizes, smarter functions, bigger quotas, as well as multiple improvements to the web UI. Today we've gone even further, announcing several updates that give BigQuery the ability to work in real-time, query subsets of the latest data, more functions and browser tool improvements.Read More

Google BigQuery is designed to make it easy to analyze large amounts of data quickly. This year we've seen great updates: big scale JOINs and GROUP BYs, unlimited result sizes, smarter functions, bigger quotas, as well as multiple improvements to the web UI. Today we've gone even further, announcing several updates that give BigQuery the ability to work in real-time, query subsets of the latest data, more functions and browser tool improvements.



Real-time analytics

BigQuery is now able to load and analyze data in real time through a simple API call, the new tabledata().insertAll() endpoint. This enables you to store data as it comes in, rather than building and maintaining systems just to cache and upload in batches. The best part? The new data is available for querying instantaneously. This feature is great for time sensitive use cases like log analysis and alerts generation.



Using it is as easy as calling the new endpoint with your data in a JSON object (with a single row or multiple rows of data).



Here's a Python example:

body = {"rows":[
    {"json": {"column_name":7.7,}}
    ]}
response = bigquery.tabledata().insertAll(
    projectId=PROJECT_ID,
    datasetId=DATASET_ID,
    tableId=TABLE_ID,
    body=body).execute()

Streaming data into BigQuery is free for an introductory period until January 1st, 2014. After that it will be billed at a flat rate of 1 cent per 10,000 rows inserted. The traditional jobs().insert() method will continue to be free. When choosing which import method to use, check for the one that best matches your use case. Keep using the jobs().insert() endpoint for bulk and big data loading. Switch to the new tabledata().insertAll() endpoint if your use case calls for a constantly and instantly updating stream of data.



Table decorators for recent imported data

This release introduces a new syntax for "table decorators". Now you can define queries that only scan a range or spot in the previous 24 hours. Traditionally BigQuery has always done a "full column scan" when querying data, while the new syntax will allow you to focus only on a specific subset of the latest data. Querying a partial subset of data with these decorators will result in lower querying costs — proportional to the size of the subset of the queried data.



With the new capabilities you can query only the last hour of inserted data, or query only what was inserted before that hour, or get a snapshot of the table at a specific time. The table decorators also work in all table related operations (list, export, copy, and so on), giving you the ability to do operations like copying a table as it was before a bad import job.



New window and statistical functions

To the previously announced window functions, we've added SUM(), COUNT(), AVG(), MIN(), MAX(), FIRST_VALUE, and LAST_VALUE(). Refer to the previous release announcement for a run through of how window functions work.



We also recently introduced the new Pearson correlation function, and showed how to find interesting and anomalous patterns in ambient sensors, or how to predict tomorrow's flight delays using 70 million flights dataset. Today's release adds new statistical functions, for even richer capabilities: COVAR_POP(), COVAR_SAMP(), STDDEV_POP(), STDDEV_SAMP(), VAR_POP() and VAR_SAMP().



BigQuery browser tool: Improved history retrieval

You can now query your history faster in the browser tool using the Query History panel: More information about the queries is being surfaced and convenient buttons have been added for common tasks.

You can try out the new UI features, table decorators, and new functions at https://bigquery.cloud.google.com/. You can post questions and get quick answers about BigQuery usage and development on Stack Overflow. We love your feedback and comments. Join the discussion on +Google Cloud Platform using the hashtag #BigQuery.



- Posted by Felipe Hoffa, Developer Programs Engineer

Whether you are a startup considering Google Cloud Platform for a new venture or an existing company who wants to breathe new life into the way you do business, we’re here to help. We have experienced solutions architects and software engineers developing architecture frameworks and adaptable solutions to help you get started on Cloud Platform.Read More

Whether you are a startup considering Google Cloud Platform for a new venture or an existing company who wants to breathe new life into the way you do business, we’re here to help. We have experienced solutions architects and software engineers developing architecture frameworks and adaptable solutions to help you get started on Cloud Platform.



Based on the strong interests we've heard from customers, we've created resources for you that focus on mobile, gaming, big data, big science, and complex web apps to start. We are providing sample applications, tools and architecture frameworks to make it as easy as possible for you to build apps and run your business on Cloud Platform.



Check out the solution scenarios and sample applications we have already published on the Cloud Platform resources page.



Here are just some of the articles we’ve already published:



Dedicated server gaming solution – In depth gaming solution, providing guidance on how to create a scalable and reliable online gaming solution

Getting started with Google BigQuery – Technical article providing a basic understanding of how Google BigQuery works, and an overview of the end-to-end process from loading data into BigQuery to visualization

Digital media asset management and sharing - Comprehensive example scenario that provides a technical deep-dive on how to use Cloud Platform to build a digital media asset management and sharing system

Also, take a look at some of the sample apps (with working source code) that we have already made available for you to download and extend:



Load balanced gaming server

Mobile shopping assistant



We have more technical articles, case studies and sample apps in the works, and we'll let you know about them as soon as they are ready.



Your first stop for all this great technical content is cloud.google.com/resources. Enjoy!



-Posted by Dora Hsu, Solutions Director

Cross-posted with the Google Developers Blog



With Google Cloud Platform, you can now get your hands on technologies that help power our infrastructure, and use it to power your own software.Read More

Cross-posted with the Google Developers Blog



With Google Cloud Platform, you can now get your hands on technologies that help power our infrastructure, and use it to power your own software.



In order to help everyone get more comfortable using Cloud Platform, we are hosting five events for developers to meet Cloud Platform senior leadership, learn to code on our stack of products, participate in code labs, and ask questions of our engineering team.



This fall, we’ll visit destinations in Europe, the Middle East and Asia:

• Tokyo (Sep. 19-20)


• Paris (Oct. 1-3)


• Berlin (Oct. 1-3)


• Tel Aviv (Oct. 7-8)


• London (Oct. 8-11)

Each stop has two days dedicated to developers. Day 1 offers a series of talks explaining why you would want to use Cloud Platform and how to build apps on it. Day 2 consists of three code labs which provide participants with an opportunity for hands-on learning using our lab setup.



To help each participant get the most out of their newly-gained knowledge, every attendee will receive a startup pack equivalent to $2,000 USD in Cloud Platform credits.



Pre-register soon, because space is limited and we expect the seats to fill fast. We will confirm your registration as soon as we can, but no later than two weeks prior to each event.



We look forward to seeing you!



-Posted by Alex Maier, Community Manager

Hadoop is a common solution for companies and research organizations taking advantage of the Big Data revolution. Many find that as their data grows, their IT infrastructure and budget cannot keep up with the storage and computational demands required.Read More

Hadoop is a common solution for companies and research organizations taking advantage of the Big Data revolution. Many find that as their data grows, their IT infrastructure and budget cannot keep up with the storage and computational demands required.



Google Cloud Platform provides a compelling alternative to purchasing and managing more servers and storage devices by providing consistent high performance virtual machines that you can pay for by the minute. To help you get started we are releasing two solution papers and two sample applications to get you up and running with Hadoop on the Google Cloud Platform. Learn more on the solution page.



Apache Hadoop, Hive, and Pig on Google Compute Engine

Have you heard about Hadoop, MapReduce, Hive, or Pig, but aren’t sure why you would use them? Or are you already running Hadoop and related tools on-premise and want to know what it will look like on Google Compute Engine? Read the solution paper, a complete guide to help you.



Managing Hadoop Clusters on Google Compute Engine

Are you already running a long-lived, mission critical Hadoop cluster on Google Compute Engine and looking for management advice? Read the solution paper for a comprehensive review on how to manage them.



Get the code and get going

Launching and managing multiple machine instances, setting up users, assigning appropriate permissions, and installing and configuring software can be a minefield that lies between you and productivity. Our sample applications get you up and running quickly.



Download or fork the GitHub projects:

• Apache Hadoop on Google Compute Engine


• Apache Hive and Pig on Google Compute Engine

- Posted by Matt Bookman, Solutions Architect





“Apache," "Apache Hadoop," "Hadoop," "Apache Hive," "Hive," "Apache Pig," and "Pig" are trademarks of the Apache Software Foundation.

Just a few weeks following our last release, we’re back with Google App Engine 1.8.4. To make it even easier for developers to build and test their applications at scale, we’re better aligning the development environment with the App Engine production environment. We’re ...Read More

App Engine 1.8.4 released

Just a few weeks following our last release, we’re back with Google App Engine 1.8.4. To make it even easier for developers to build and test their applications at scale, we’re better aligning the development environment with the App Engine production environment. We’re updating the development application server to perform more like the High Replication Datastore, so now it defaults to using an eventually consistent storage model.



We also updated our support for Dynamic Web Projects in Eclipse to better support Google Cloud Endpoints and App Engine Backends as well as addressing a number of bugs users have reported. Try it out — download the latest GPE release.



To see what else is included in 1.8.4, please do check out the release notes.



Differential snapshots for Persistent Disks

Cloud Platform makes it easy to create a snapshot of a Google Compute Engine persistent disk and keep that snapshot copy stored in Google Cloud Storage for backup, disaster recovery and migration purposes. More importantly, you can boot a new virtual machine in seconds using that persistent disk snapshot anywhere in the world. Now, with differential snapshots, Cloud Platform only has to store the data that has changed since the last successful snapshot resulting in significantly faster snapshot times with less Cloud Storage usage (meaning less cost to you!). As a result, you can take snapshots more regularly and have better protection for critical data.



Compute Engine load balancing — now with a UI!

Following the recent announcement of Google Compute Engine load balancing, we've added it to Cloud Console. The new interface gives you an easy way to administer your load balancing.

To find it, log in to Cloud Console and look for "Load balancing" on the left menu.



We’d love to hear your feedback in the comments. If you have specific technical questions, check out our Google Groups or find us on Stack Overflow.



-Posted by Jessie Jiang, Director of Product Management

BigQuery is transforming the ways in which we work with massive amounts of data. Our partners have created amazing tools to make that process even more streamlined: Visualizing, slicing, and dicing. Working with your intuition and these tools, you can discover surprising new insights, analyzing terabytes of data in mere seconds. What's still a challenge is feeding this intuition, discovering where to look for insights.Read More

Editor's note: Join Felipe Hoffa and Michael Manoochehri tomorrow, September 6th at 11:00AM PST on Google Developers Live where they'll discuss correlation with BigQuery.

Google BigQuery is designed to make it easy to analyze large amounts of data quickly. We are always looking into how to make BigQuery even more powerful, so today we'll introduce a feature that we couldn't wait to share with you: Pearson correlation.

BigQuery is transforming the ways in which we work with massive amounts of data. Our partners have created amazing tools to make that process even more streamlined: Visualizing, slicing, and dicing. Working with your intuition and these tools, you can discover surprising new insights, analyzing terabytes of data in mere seconds. What's still a challenge is feeding this intuition, discovering where to look for insights.



The new CORR() function in BigQuery is a powerful tool for your intuition process: Which variables are similar, or have surprising behaviors? Can you rank these surprising behaviors? What are the best variables to predict the future?



Let's look at the data we collected at the Data Sensing Lab at the Moscone Center for Google I/O. We had multiple sensors in multiple rooms, collecting temperature, humidity, noise and other data during these 3 days.



To start, we can look at what rooms' temperature behaved in a similar way during the 2nd and the 3rd day:

SELECT CORR(a.data, b.data) corr, a.room room, count(*) c
FROM (
  SELECT
    TIME(USEC_TO_TIMESTAMP(INTEGER(Timestamp / 60000000) * 60000000)) time, AVG(DATA) data, room
  FROM [io_sensor_data.moscone_io13]
  WHERE
    DATE(USEC_TO_TIMESTAMP(Timestamp- 8*60*60000000)) = '2013-05-16'
    AND sensortype='temperature'
    GROUP EACH BY time, room) a
JOIN EACH (
  SELECT
    TIME(USEC_TO_TIMESTAMP(INTEGER(Timestamp / 60000000) * 60000000)) time, AVG(data) data, room
  FROM [io_sensor_data.moscone_io13]
  WHERE
    DATE(USEC_TO_TIMESTAMP(Timestamp- 8*60*60000000)) = '2013-05-17'
    AND sensortype='temperature'
    GROUP EACH BY time, room) b
  ON a.time=b.time AND a.room = b.room
  GROUP EACH BY room
HAVING
  corr IS NOT NULL
  AND c > 800
  ORDER EACH BY corr DESC

corr	room	c
0.9387693711	sf desk 1st floor	1331
0.8488553811	chrome east	1418
0.8423597116	chrome hobbit	1372
0.8162574011	chrome west	1401
0.7696065852	chrome north	1374
...	...	...
-0.1048712561	Room 1	1390
-0.1508345595	keynote crowd	1358
-0.5467798237	android east	1402

This table says that many rooms behaved in a similar way during both days. For example, a room inside the main exhibition area:

We see that the temperature rose during the day, in a very similar way. A/C helps a lot with this. Meanwhile in the main reception area the temperature dropped considerably when the doors opened:

Even though the base temperature was considerably higher on Day 2, CORR() finds out that the behavior was very similar.



The room 'android east' shows a highly negative correlation. We can visualize why:

Turns out something was wrong with the sensors there a few hours during Day 2, and went to 0 during Day 3. It's too late now to go and fix them, but we can keep this as a good example of how CORR() can help us set alarms and notify on unusual events.



We can repeat this experiment with other dimensions, like humidity or noise. But we can go further, and find out what dimensions correlate with each other.

SELECT CORR(a.data, b.data) corr, a.sensortype a_sensortype, b.sensortype b_sensortype, a.room room, count(*) c
FROM (
  SELECT
    TIME(USEC_TO_TIMESTAMP(INTEGER(Timestamp / 60000000) * 60000000)) time,
    AVG(DATA) data, room, sensortype
  FROM [io_sensor_data.moscone_io13]
  WHERE
    DATE(USEC_TO_TIMESTAMP(Timestamp- 8*60*60000000)) = '2013-05-16'
    AND sensortype != 'pressure' AND sensortype != 'altitude' AND room != 'None'
    GROUP EACH BY time, room, sensortype) a
JOIN EACH (
  SELECT
    TIME(USEC_TO_TIMESTAMP(INTEGER(Timestamp / 60000000) * 60000000)) time,
    AVG(data) data, room, sensortype
  FROM [io_sensor_data.moscone_io13]
  WHERE
    DATE(USEC_TO_TIMESTAMP(Timestamp- 8*60*60000000)) = '2013-05-16'
    AND sensortype != 'pressure' AND sensortype != 'altitude' AND room != 'None'
    GROUP EACH BY time, room, sensortype) b
  ON a.time=b.time AND a.room = b.room
  WHERE a.sensortype > b.sensortype
  GROUP EACH BY room, a_sensortype, b_sensortype
HAVING
  corr IS NOT NULL
  AND c > 800
  ORDER EACH BY corr DESC

corr	a_sensortype	b_sensortype	room	c
0.9452553258	temperature	humidity	android east	1436
0.9230651809	temperature	gas	sf desk 1st floor	1425
...	...	...	...	...
0.7949466754	temperature	mic	maps	1434
...	...	...	...	...
0.0013869924	temperature	light	Room 12	1436
...	...	...	...	...
-0.9396195218	temperature	humidity	accesability	1341
-0.9564087934	temperature	humidity	chrome lounge	1135
-0.9925864599	temperature	humidity	sf desk 1st floor	1437

We can visualize how the air quality in the reception improved, once the doors were opened - as the drop in temperature shows:

Or how the temperature rose at the same time as the volume levels, in the Maps area:

Temperature and light with almost no correlation in Room 12 calls our attention, as we got some very bad measurements in that room:

Temperatures of over 2000? Something went wrong.



At the end of the table, it's a surprise that while humidity had a high negative correlation to temperature in many sectors, it had a high positive correlation in 'android east'. First we can look at the normal behavior in the 'chrome lounge sector':

And the unusual behavior at 'android east':

Turns out there is a highly unusual positive correlation because both sensors went to 0 at the same time.



That's the nature of real data. It's a hard job to keep it accurate and clean. CORR() is a tool that can also call our attention to what is going wrong.



We can go much further with this: What's the relationship between rooms? What sensors in which rooms can help us predict what will happen an hour later on a different room? For this and much more tune in for an upcoming Google Developers Live session where we'll discuss how we built this queries, and how to go beyond. Stay tuned!



-Posted by Felipe Hoffa, Developer Programs Engineer

Google Cloud Platform is used around the world to create and deploy solutions that are affordable, scalable and easy to maintain. These qualities empower developers worldwide to solve local and global challenges with cloud technology .Read More

Google Cloud Platform is used around the world to create and deploy solutions that are affordable, scalable and easy to maintain. These qualities empower developers worldwide to solve local and global challenges with cloud technology .



This sets the stage for the very first Google Cloud Platform Developer Challenge. We’re inviting you to build locally relevant web applications that solve real world problems. You will have the opportunity to “WOW” the world with your awesome web application built on Google App Engine using Google APIs like the Google+, YouTube and Maps APIs.

The competition starts September 4th, 2013 with the first round of submissions starting on the October 22nd, 2013. The categories for this challenge are

• Enterprise/Small Business Solutions , Education, Not for Profit


• Social / Personal Productivity/Games / Fun

Submissions will be accepted from any one of six regions across the world:

• Latin America


• Sub Saharan Africa


• South East Asia


• Middle East & North Africa


• India


• Rest of the World

We will be working in tandem with Google Developer Groups (GDGs), Google Business Groups and Google Student Ambassadors to put up and organize events around the challenge. Ensure you connect with your local communities for more information about these meetups.



Do you want to start hacking out a cloud based web app now and you need pointers and hints? If yes , then the Google App Engine tutorial is a good place to start.

The Form+ team, one of the winners of last year's Google Apps Developer Challenge

If your application is judged the best in your region, within one of the two categories mentioned above, you could win a prize of $20,000! We are devoted to raising the next generation of information technology experts as well as encouraging diversity among software developers. In line with this, we have special prizes for all-female teams -- $2,000 and $1,000 dollars for ‘all-student teams’ that make the second round. Any ‘all female student team’ that makes the second round will receive a prize of $3,000. There are also prizes available for those who are interested in mentoring contestants.



Please read through the terms and conditions on the Google Cloud Platform Developer Challenge page and follow the hashtag #gcdc on Google+ for any additional updates.



What are you waiting for? Register your gcdc2013-.appspot.com - apply for $2,000 worth of Google Cloud Platform credit at cloud.google.com/starterpack using the promo code: gcpdc-in, and start hacking with Google Cloud Platform!



-Posted by Chukwuemeka Afigbo, Program Manager