At Scalr, we’ve been happy customers of Amazon’s infrastructure service, EC2, since 2007. In fact, we’ve built our tools for EC2 because we saw an opportunity to leverage its flexibility to help AWS customers easily design and manage resilient services. But as competitors spring up, we always test them to see how they compare, especially in regards to io performance.

On a warm June day in San Francisco, the Scalr team attended Google I/O 2012. Google was rumored to be launching a EC2 competitor, which we were interested in for our multi-cloud management software. It launched. And boy did it sound good. You see, EC2 and GCE offer pretty much the same core service, but Amazon has been plagued by poor network and disk performance, so Google’s promise to offer both higher and more consistent performance struck a real chord.

Not ones to be fooled by marketing-driven, hyped-up software, we applied for early access and were let in so we could start testing it ourselves. Once we got in, we felt like kids in a candy store. Google Compute Engine is not just fast. It’s Google fast. In fact, it’s a class of fast that enables new service architectures entirely. Here are the results from our tests, along with explanations of how GCE and EC2 differ, as well as comments and use cases.

A note about our data: The benchmarks run to collect the data presented here were taken twice a day, over four days, then averaged. When a high variance was observed, we took note of it and present it here as intervals for which 80 percent of observed data points fall into.

API

First off, GCE’s API is beautifully simple, explicit and easy to work with. Just take a look at it. Their firewalls are called “firewalls,” vlans are “networks,” and kernels are “kernels” (AKIs, anyone?). Anyone familiar with Unix will feel right at home.

Fast boot

Second, VMs are deployed and started with impressive speed (and we’ve extensively used 10 clouds). It routinely takes less than 30 seconds to login as root after making the insert call to launch a VM. As a reference point, this is the amount of time it takes AWS to get to the running state, after which you still need to wait for the OS to boot, for a total of 120 seconds on a good day, and 300 on a bad one (data points taken from us-east-1).

Boot times are measured in seconds.

We don’t know what sort of sorcery Google does here, but they clearly demonstrate engineering prowess. That’s 4-10x faster.

Volumes

Those of you familiar with Amazon’s EBS volumes know that you can attach and detach volumes to any instance, anytime. On GCE, you can’t (at least not yet). This precludes you from switching drives to minimize downtime: attaching a volume on a running server allows you to skip the boot and configure stages of bringing a new node up, which is useful when promoting an existing mysql slave to master and you just need to swap out storage devices.

While GCE’s “disks” (as they call them) have that one disadvantage, they offer some unique advantages over Amazon volumes. For example, disks can be mounted read-only on multiple instances, which makes for more convenient fileserving than object stores, especially for software such as WordPress (see disclosure) or Drupal that expect a local filesystem. Disks are really fast too, and don’t seem to have the variable performance that plagued EBS before the introduction of Provisioned IOPS. See for yourself in the following benchmarks.

As you can see, GCE and EC2 are equivalent on reads, but GCE is 2-4x faster on writes.

Read/write speeds are measured in MB/s. Higher numbers mean faster throughput.

Network

A short note about multi-cloud. I’m talking here about services that span multiple clouds, such as replicating a database from us-east-1 to us-west-1, for disaster recovery or latency-lowering purposes, not the multi-cloud management capabilities widely used in the enterprise. I believe that first kind of multi-cloud is a myth driven by the industry’s less tech-savvy folks. I’ve seen too many people attempt it unsuccessfully to recommend it: what usually happens is the slave database falls behind on the master, with an ever-increasing inconsistency window, because the load on the master exceeds the meager bandwidth available between master and slave. Our friends at Continuent are doing great work with Tungsten to accelerate that, but still.

Google’s network is so fast, however, that this kind of multi-cloud might just be possible. To illustrate the difference in speeds, we ran a bandwidth benchmark in which we copied a single, 500 Mb file between two regions. It took 242 seconds on AWS at an average speed of 15 Mbit/s, and 15 seconds on GCE with an average speed of 300Mbit/s. GCE came out 20x faster.

Higher bandwidth is better and means faster up and downlinks.

After being so very much impressed, we made a latency benchmark between the same regions. We got an average of 20ms for GCE and 86ms for AWS. GCE came out 4x faster.

Lower latency is better and means shorter wait times.

This might allow new architectures, and high-load replicated databases might just become possible. Put a slave in different regions of the US (and if/when GCE goes international, why not different regions of the world?) to dramatically speed up SaaS applications for read performance.

Of course, unless Amazon and Google work together to enable Direct Connect, bandwidth from GCE to EC2 will still be slow. I also hear that Amazon is working on creating a private backbone between regions to enable the same use cases, which would be an expected smart move from them.

Multi-region images

We’re not quite sure why AWS doesn’t support this, but images on GCE are multi-region (“multi-zone” in their terms), that is to say when you snapshot an instance into an image, you can immediately launch new instances from that image in any region. This makes disaster recovery that much easier and makes their scheduled region maintenance (which occurs a couple of times a year) less of a problem. On that note, I’d also like to add that it forces people to plan their infrastructure to be multi-region, similar to what AWS did for instance failure by making local disk storage ephemeral.

So should you switch?

AWS offers an extremely comprehensive cloud service, with everything from DNS to database. Google does not. This makes building applications on AWS easier, since you have bigger building blocks. So if you don’t mind locking yourself into a vendor, you’ll be more productive on AWS.

But that said, with Google Compute Engine, AWS has a formidable new competitor in the public cloud space, and we’ll likely be moving some of Scalr’s production workloads from our hybrid aws-rackspace-softlayer setup to it when it leaves beta. There’s a strong technical case for migrating heavy workloads to GCE, and I’ll be grabbing popcorn to eagerly watch as the battle unfolds between the giants.

Sebastian Stadil is the founder of Scalr, a simple, powerful cloud management suite, and SVCCG, the world’s largest cloud computing user group. When not working on cloud, Sebastian enjoys making sushi and playing rugby.

Note: Data scientists from LinkedIn, Continuuity, Quantcast and NASA will talk about their hardware and software stacks at our “guru panel” at Structure:Data next week, March 20-21, in New York City.

Disclosure: Automattic, maker of WordPress.com, is backed by True Ventures, a venture capital firm that is an investor in the parent company of this blog, GigaOM. Om Malik, founder of GigaOM, is also a venture partner at True.

Update: This story was updated at 7:34 a.m. PDT on May 15, 2013 to note that Scalr is working on a new set of benchmarks and will publish those results soon.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Migrating media applications to the private cloud: best practices for businesses
• Takeaways from the second quarter in cloud and data
• Metered IT: the path to utility computing

Few, if any, of the hardware components will surprise those that follow Android as the leaked specifications proved true. The 7-inch slate is built on Nvidia’s low cost Kai platform, using the Tegra 3 system on a chip: a quad-core processor with a low powered fifth core and 12 graphics cores. The 1280 x 800 resolution screen looks good and has wide viewing angles. There’s no rear camera sensor, but the front facing camera works well for video chats. So far, the package is peppy in my limited testing.

Part of the reason for that is surely the new software: Nexus 7 is the first device to ship with Android 4.1, also known as Jelly Bean. This too was introduced at Google I/O with a number of refinements and new features that help make the platform more useful with a smoother user experience to boot. Google outlines the many new features on its developer blog here.

I’m already impressed by Google Now, which is part Google Search, part personal assistant, and all controlled by voice. Google Now is contextual in that it learns about you over time and begins personalizing information, sometimes in advance. For example, the software will quickly learn where your home is and when you leave work it can proactively inform you of the commute time. In addition, if there is a way to route around any traffic issues to get you home sooner, Google Now will do that for you as well.

Also included in Jelly Bean are welcome notification improvements. In the past, you could only swipe away a group of unread emails, but now you can wipe individual messages away from your notifications or take immediate action on them. And developers can allow for up to three actions on any notification so you may be able to respond to a Twitter message right from the notification pane, for example.

Jelly Bean also adds some core system optimizations, which Google calls “Butter”, to speed up the overall experience. I’ve already installed Android 4.1 on my Galaxy Nexus, plus I’m using it on the Nexus 7, and see the difference. I wouldn’t say that Android becomes twice as fast as before with Jelly Bean, but it’s a noticeable speed boost for navigation, program switching and all around use.

Unfortunately for those considering to use Jelly Bean on a brand new Galaxy Nexus, there may be a wait. Late in the week, Apple won a preliminary injunction against Samsung for the handset. The unlocked GSM handset dropped in price, now at $349,  just a few days before the injunction.

This is less of an issue for Samsung, however, than it is for Google. Samsung’s Galaxy S III, which could also be targeted for patent violation, is surely a better seller. Google is trying to crack the direct handset sale market with the Galaxy Nexus, so I suspect that the company will finally get involved in the legal spats to help out its hardware partners as a result.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Tablet market to hit over 377 million units by 2016
• CES 2012: a recap and analysis
• The fourth quarter of 2012 in mobile

Actor/geek Wil Wheaton started to see the results almost immediately, posting late Wednesday afternoon:

Did G+ roll out some new “invite everyone you follow to an event” thing? My feed is completely overwhelmed with “everyone’s invited to XXXXX event” notices, and I’m having a hard time actually seeing posts from people I’m following. … Is there a way to opt-out of event invites from people I don’t follow or have circled? Is this yet another thing Google rolled out without thinking it through clearly?

A little later, he was getting a better understanding of the new feature but it was too late:

It’s too bad that I’ll never use it, and will kill it with fire, because Google has, yet again, made a product that may be useful and cool, but forced it upon users without giving users any control over how invasive it is.

Wheaton not only has a large following, he’s an active poster and checkers his Google+ feed every half hour or so when he’s online. He likes Google+ and is just as frustrated by what he sees as Google’s failure with it as he was with this feature gone awry:

Instead of talking about how cool this thing is, and how excited we are to use it, a significant number of G+ users — and people like Linus Torvalds, who are way more influential than I am are among them — are talking about how annoyed they are and how much they hate it.

Is that the rollout that Google was hoping for? Has Google learned nothing from doing things like this in the past? I’m starting to believe that this isn’t an accident, or poor planning, but by design; I just can’t figure out why.

Author John Scalzi was irked enough to post around midnight Wednesday:

My newest circle is “People who send me a ‘test’ Google Events Invite, and must therefore be torn apart by ravenous bears.” Try not to be in it. The bears are getting tired.

By Thursday, Wheaton was trying again, providing examples, to back up his contention that only people in your own circles should be allowed to send you Events. Among them: unwanted invitations and messages about the Guild Wars 2 release date. The last straw was someone who set up a fake event with Wheaton’s name in it, prompting him to warn “the system as it’s set up right now is ripe for abuse.”

That finally drew a chagrined response in Wheaton’s comments from Vic Gundotra, the Google SVP of engineering responsible for Google+:

“We are doing exactly what you requested. We should have contemplated and anticipated how people would abuse this and how painful this could be for celebrities with large followings.

We have pushed a number of fixes yesterday (some were bug fixes that showed up at scale). Expect more fixes today. Sorry for the trouble Will.”

The right response, although Gundotra shouldn’t have limited it to “celebrities” and should have been more clear about the fixes and bugs.

An exasperated Robert Scoble was still dealing with the aftereffects late Thursday afternoon when he broadcast on Google+ and Twitter that Google+ Events was “The worst social launch ever. WTF were you thinking Google?” He called out Gundotra directly, adding a new twist:

Not only did it spam the crap out of my notifications and my Google+ events page but it added events — hundreds of them — onto my calendar.

My calendar is MINE. Not yours. You should NEVER put anything on it that I don’t approve of.

Scoble said declining the events didn’t make them disappear, requiring hundreds of entries to be deleted by one by one. It’s not the end of the world but it’s the kind of tech time sink that can frustrate to the point of a primal scream.

Google also is showing an Events opt-in screen on Google Calendar. I was in a rush when I first saw it and decided to adhere to a rule I’ve been trying to follow: don’t opt in if you don’t have time to read the fine print.

Even now, it’s difficult to see from the splash screen or the “learn more” page just how much noise you might be agreeing to if you opt in.

If you look at events through Google+ instead of your Google calendar, you can start to figure it out by choosing settings and scrolling around a lot. There is still no simple or direct way to set parameters specifically for events, as far as I can tell.

Google needs Google+ Events to work for a number of reasons: it encourages interaction between features like Google+, calendar and Gmail; it’s social so should bring more Google+ users together; it encourages photo sharing; and it should bring new users in.

If Google can manage the parachute-bike-rappel routine that wowed the Moscone Center crowd, someone there ought to be able to figure out how to introduce new features without alienating its users.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• The 2013 task management tools market
• How consumer media will change in 2013
• Social 2013: The enterprise strikes back

To use offline document editing in Google Docs, users will have to enable in it the applications settings while online. Docs will then download local copies of all documents for editing without a connection. Once that step is complete, edits can take place offline and when Docs senses that a connection is available, it will automatically sync the changed documents with the files in Google’s cloud.

Of course, Google Docs is going away in favor of Google Drive: Google recently announced the change in an effort put file storage and editing capabilities under one name. But there’s more to Google Drive than just the name and iOS device owners can start to take advantage of Google’s storage service with a native iOS app for Drive. There is one caveat: It’s just a file viewer for now.

All Google Drive users gain a nifty new feature: A vastly improved search function. On stage at Google I/O, the company demonstrated this and it was impressive. About two dozen PDF files of scanned images were shown in a Google Drive account and none were labeled. Yet one out of the bunch was a shipping receipt.

A quick search for “certified mail” pulled up the proper image thanks to optical scanning recognition of the file contents, making Google Drive much more useful and easier to use: Why label what you don’t have to?

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• The evolution of consumer-media cloud storage
• Analyzing the wearable computing market
• How scribbling on an iPad makes your work life easier

The new browser brings cross-device syncing to iOS, making it possible for users of the Chrome desktop browser to access any of the tabs they have opened on their PC on their iOS mobile device as well. Chrome also synchronizes browser history, making it possible to use the back button to access previously opened sites.

Google’s Sundar Pichai said earlier during the keynote that mobile will be the biggest growth market for Chrome. Three billion people already access the web, he said, but half of the world’s population still have yet to come online. “And those we reach via mobile,” he said, with Chrome being key to these effort. “Chrome is the most popular browser in the world, globally,” he declared.

Update: The release coincides with Google making Google Drive available on iOS. More on this in our live blog.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Analyzing the wearable computing market
• Carrier IQ and the continued erosion of operator trust
• The future of mobile: a segment analysis by GigaOM Pro

With an epic Day One of Google I/O in the rearview mirror — a day that included a major new release to Android, a new Google-branded tablet and home entertainment device, and a wacky high-wire stunt act involving sky divers and Google Glasses — Day Two might find itself with a hard act to follow. However, Google is expected to focus much more on the Web during the second day of its Google I/O show, with news expected from its Chrome and Chrome OS divisions as well as a new cloud service that you first heard about here at GigaOM.

Just like yesterday, we’ll be providing live coverage of the event here, with frequent updates over the course of the morning’s keynote session as well as links to our coverage throughout the day. The live updates will begin (hopefully) around 10 a.m. PT, assuming Google irons out the kinks in the crowd-control measures for the Google I/O keynotes as to permit everyone to enter the third floor of the Moscone Center in a timely fashion.

During the morning events, this page will refresh automatically every five minutes with the latest updates starting at 10 a.m., and we’ll plug in links to our coverage as stories emerge.

Update: Several sessions for developers remain over the rest of today and tomorrow, but that’s it for the keynote presentations. Here are the stories that have emerged from Google I/O on Thursday:

• Google releases Chrome for iPhone and iPad
• Taking on Amazon, Google launches EC2 rival Compute Engine
• Google Docs goes offline while Drive comes to iOS

9:53 a.m.: Either there are a lot less people here today for Day Two of Google I/O, or Google figured out how to manage huge crowds, because getting onto the third floor of Moscone Center was a breeze today. The hall is filling, and we’re waiting for Google to start.

10:07 a.m.: Google’s kicking off the second day with a deafening countdown. We’re all awake now. Vic Gundotra is going to be the master of ceremonies for today’s event as well. Vic’s rocking the baby blue Google Glasses on stage today.

10:11 a.m.: As expected, this is going to be a Chrome and Chrome OS day, which means that Sundar Pichai, last seen at Structure 2012 last week, is out to talk about what’s new. He says he’s going to focus on the cloud and Chrome OS, as well as services for developers. If you’ve been reading Om’s stories lately, you’ve got a pretty good idea what he means by that last one. Network connectivity is exploding, he said, as mobile devices become the primary way of getting onto the Internet for lots of people.

10:14 a.m.: There are 310 million people using Chrome right now, Pichai said. He’s claiming that Chrome is the most popular browser in the world, which is a bit controversial an assertion at the moment given the way recent market-share stats from StatCounter were presented.

10:17 a.m.: Brian Rakowski, vp president for Chrome, has taken the stage to talk about how Chrome can be used across multiple devices. Most people switch between home and work devices every day, and Chrome settings can of course be synced across different computers. Nothing ground-breaking so far, which makes it really easy to tell where the Google employees are seated in the keynote hall when wild cheers and applause greet the notion of synced bookmarks.

10:20 a.m.: Rakowski also demonstrates how this feature works on mobile devices running Chrome, which is still a new notion of course, but Chrome will be the default browser on the Nexus 7 tablet unveiled yesterday now that it is ready to come out of beta.

But now we’re getting something new: Google is releasing Chrome for the iPhone. It will arrive in the App Store later today.

10:24 a.m.: There will also be a version for the iPad. It works pretty much the same as the regular version of Chrome, with synced settings between different devices and “icognito mode.” “No matter which device you are using, we are working really hard across all software platforms,” Pichai says upon returning to the stage.

10:27 a.m.: Pichai is now shifting into Google Apps mode, talking about the various apps that Google has introduced since 2004 and Google’s pitch to get more businesses using Google Apps internally. Five million businesses have signed up for Google’s application services, he said. Enterprise computing has switched from maximizing individual productivity to fostering collaboration among employees, which means that new computers and new services are needed. And now, it’s time for a Google Apps commercial!

10:31 a.m.: I actually kind of like the Hall and Oates one, if you haven’t seen that yet. We’re now going to get an update on Google Drive from Clay Bavor, director of product management for Google Apps. Starting to sense a theme: Google Drive will now be available for iOS and Chrome OS. Bavor moves into a demonstration of Google Drive on the iPad, showing how you can search for documents and files stored in remote locations.

10:35 a.m.: Google Drive for iOS will be available later today. Bavor moves over to a Chromebook, which is a little different because Chromebooks don’t have internal storage, so Google Drive is really like the file system for Chrome OS. He also demonstrates Google Docs on the Chromebook, and finally solves one of the big issues with Chrome OS: Google Docs will now work offline.

10:40 a.m.: There’s a new SDK available for Google Drive, which allows third-party developers to integrate Google Drive into their apps. The guys from HelloFax (an initial Google Drive partner) get a shout-out in the keynote, and LucidChart is also used as an example. Developers report that Google Drive integration makes for more engaged users than those developers see from folks coming directly to their apps, Bavor claims. A second version of the software is being released today, with explanatory sessions to follow.

10:42 a.m.: Pichai is back to talk about Chrome OS. He’s focusing on the speed of the second-generation Chromebooks, which were much faster than the initial version, as well as the automatic updating of the software. Chromebooks are coming to physical retail stores through a partnership with Best Buy in the U.S. and Dixons in the U.K. A new lineup of devices seems planned for the holiday shopping season.

10:47 a.m.: Urs Holzle is out to talk about cloud applications and web developers. He starts out with some updates on Google App Engine, some of which Barb Darrow covered yesterday. “Many of you have said you want even more options,” he said, noting that folks are calling for “many virtual machines.” He announces Google Compute Engine, a service that will let developers run their apps on Google’s cloud.

10:50 a.m.: Urs is showing how the Institute for Systems Biology, a Compute Engine beta tester, used the service. This was a pretty complex app called Genome Explorer, and it could take up to 10 minutes to process tasks on the group’s in-house infrastructure. On Google’s service, those tasks were reduced to seconds.

10:54 a.m.: Google is selling Compute Engine on “scale, performance, and value,” taking the first clear shot at Amazon’s rival service. It’s open now in a limited preview: “10,000 cores, I think that’s really cool. But you know what’s really cool,” Holzle asks with a smile? Google can scale higher if needed, referring to a counter that has been running behind Holzle’s presentation. It stops at 771,886 cores, which are apparently available to app developers. The Institute for Systems Biology demonstration pops when given 600,00 cores to use, as compared to the 100,000 it was using before.

10:59 a.m.: Pichai is back. He’s talking about games at the moment, demonstrating how games can be streamed online. Of course, they demonstrate a shooter. The game takes advantage of Native Client on a Chromebox to deliver the performance without crushing the internet connection. He’s running down the different games and other apps available in the Chrome App Store. Like Docs, Google is working to make other Chrome apps available offline and deliver more full-screen apps.

11:03 a.m.: Two representatives from Cirque du Soleil have come out on stage, talking about how they have been trying to create the same experience of their live shows on the web. We’re getting a demo of the web app that they created: it’s sort of like playing a game with Cirque du Soleil characters, only it doesn’t cost $139 a ticket.

11:06 a.m.: Everything was created with standard web technologies, according to the group, which impresses the developers in the crowd. It’s a very realistic simulation, which has been very hard (if not impossible) to do without dedicated hardware for years. This is still a work in progress, but it’s a nice endorsement for web technologies at the heart of things like Chrome OS.

11:07 a.m.: While I’m busy not paying attention to this Chrome OS commercial currently airing on stage, check out Om Malik’s in-depth look at Google Compute Engine.

11:12 a.m.: We’re all getting Samsung Chromeboxes! The roar from the crowd is a little less deafening then it was yesterday for the Galaxy Nexus and Nexus 7 tablets. But we’re not done: Sergey Brin is back with a live video from the roof of the Moscone Center, where yesterday Google pulled off a tech conference demo for the ages of its Google Glass project. Brin is demoing the Glasses and they are apparently going to do this again.

11:16 a.m.: Brin makes sure to note that Google coordinated this stunt with the city of San Francisco and the FAA. They’ve got more cameras in place today and they’re going to show this again as a Google Hangout, with Brin’s Glasses giving the perspective from the roof and Glasses on the jumpers showing the view on the way down. We’re waiting for the jumpmaster to give the all-clear.

11:20 a.m.: Google brought out special radios for the stunt today, which apparently will improve connectivity to the Glasses. Brin is giving a play-by-play from the roof, as opposed to his perch onstage yesterday. The jumpers all made it down safely, and the picture was a little clearer today because of the improved connectivity. A little anti-climactic, perhaps, but still pretty amazing.

11:23 a.m.: Brin is walking us through a bit of the behind-the-scenes stuff, but we’re otherwise getting basically the same stunt as yesterday. A poor camera guy just walked into a pole on the roof of the building, which got a sympathetic groan out of Sergey. Perhaps most interesting at this point is the addition of shades to Google Glasses.

11:27 a.m.: They basically repeated the same whole stunt from yesterday, with bikers and rappelling and all. Not sure what the point of doing that again was, but if you haven’t seen the video from yesterday, it’s worth checking out. That’s going to be all for today’s announcements, thanks again for hanging out with us.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• A near-term outlook for big data
• Dissecting the data: 5 issues for our digital future
• NewNet Q2: Google closes the quarter with a bang

• 400 million active Android devices worldwide. Around this time last year, there were 100 million active Android devices.
• Google is activating 1 million Android devices daily.
• A million NFC-enabled devices are being activated every week.
• South Korea and Japan saw 200 percent year over year growth in activations of Android devices.
• India, Thailand and Brazil saw 400 percent increase in activations on a year-over-year basis.
• There are 600,000 Android apps.
• There have been 20 billion app installs to date.
• Two-third of Play revenue comes from outside of the US.
• Google has more than 4 million books in the Play store.
• YouTube has 800 million monthly users.
• There are 350 million mobile devices with YouTube installed on them.
• Twitter users share 700 YouTube videos every minute.

And now for the Google+ numbers.

• 250 million accounts.
• 150 million monthly active Google+ users.
• 75 million daily active users.
• Apparently daily actives spend more than 12 minutes per day.

Here’s our recap of Day One as it unfolded.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• The state of cross-platform media measurement
• Connected world: the consumer technology revolution
• NewNet Q3: Facebook remakes headlines in social media

With the update, YouTube channels and subscriptions appear on the left side of a mobile device screen, highlighting the content you’re most likely interested in. You can still search for videos or catch the hottest movie short of the day, but for those that follow certain channels or video producers, this makes it easier to stay in tune.

Also included in the updated version of YouTube is a method to pre-load videos from your Subscriptions or Watch Later list into cache memory. You can’t watch them offline — Google says you’ll still need a connection to view the content — however, this should eliminate much of that spinning circle when the video file loads. Since a good portion of the video is local, your connection can grab any remaining parts of the video while you watch.

This is all good news for Android users who are YouTube addicts, but there’s a somewhat expected downside as well: The updated YouTube features are for Android 4.0 devices or better. That means unless you have a Google Nexus S or Galaxy Nexus phone — or one of the other handsets that have recently seen the Android 4.0 upgrade — you’re still stuck with that spinning circle on a slow connection. Don’t turn that dial though: Google says it will bring these features to older Android versions in the coming months.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Got a Cable Subscription? There’ll Be an App for That
• Controversy, courtrooms and the cloud in Q1
• Netflix may suffer from limited mobility

Their descent was streamed in real time and culminated in them landing on the roof of the Moscone center, where they handed off a package to some mountain bikers, only to have it handed to guys climbing down the face of the building. Words don’t really do it justice, so you might want to check the video out yourself. However, the big show couldn’t really conceal that Google had little news to offer about the actual glasses.

Check out the Project Glass demo from Google I/O:

Google did start taking pre-orders on site, offering developers in attendance of the conference a chance to put their name down on a priority list for the device. But even those pre-orders won’t likely ship till early next year. Brin said that the version they’re going to make available to developers will be rough around the edges, but that the company is trying to get it into the hands of developers as early as possible. Developers that are willing to pay a premium, one should add: The first version of Glass, called “Explorer Edition”, will cost a whopping $1500. There’s no word on how much glass will set back the average consumer when it becomes more widely available, or when this will happen.

But the biggest omission during Wednesday’s Glass presentation was any peek at the current Glass UI, or any of the features it will offer end users, save for one or two: Glass will be able to record photos and video, and users will be able to share these via Google+ and also join Google+ Hangouts.

That’s remarkably less than what the company shared when it first publicly unveiled Glass in April. Back then, Google released a conceptual video that was meant to show off some of the ideas behind Glass. The video featured an augmented reality experience, letting the wearer of Glass know about interesting spaces he was visiting, notifying him of the location of his friends and allowing him to answer phone calls without distracting too much from the real world.

But ever since this first glimpse, Google has only been showing off the device’s photo and video capture capability. Brin joked on stage Wednesday that it would be hard to show off the Glass UI because it might require him to wear two glasses at the same time. Of course, the Glass experience is far from ready – and that goes far beyond the UI.

Even with as little as Google has shared so far about the project, it’s becoming obvious that Glass will offer a completely different computing experience. It’s taking computing away from the screen, and into the real world. That poses numerous challenges, as it requires Google to completely rethink the way it has been presenting information. Its new Google Now mobile search takes a first step into this direction, away from making information searchable, to more instinctively surfacing things you want to know in any given situation. Of course, there are also privacy and basic safety implications – things Google apparently wants to work out before giving us a closer look at Glass.

It makes perfect sense for the company to involve developers in this process – but it’s less certain what developers who commit to pre-order are getting out of it. Will they be able to run their own apps on Glass? Will Glass work in concert with other Android devices? Will it lead to the birth of a new generation of online services that combine location with image recognition and other types of sensory input? We just don’t know yet, and Google won’t tell.

But judging from the lines at the pre-order table, this doesn’t seem to stop developers. We’ve heard that people pre-ordering the device now won’t have to pay until it actually ships next year – but nonetheless, Google is asking developers to take a leap of faith for Project Glass. Maybe the whole jumping out of the airplane was more than just a stunt, after all.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Analyzing the wearable computing market
• How emerging technologies will influence collaboration
• Carrier IQ and the continued erosion of operator trust

There was nothing earth-shattering, but with more than half of the market, Google just needs to keep refining the platform and staying competitive with Apple. A bigger question will be how long it will take for existing Android users to get 4.1, since only about 7 percent of Android users are on Ice Cream Sandwich.

The update does help Android compete better with iOS, bringing some cool features that stack up well against Siri. It also makes Android more of a one-stop shop for users, who don’t have to turn to other apps in some cases.

Here’s a rundown of the biggest improvements:

The biggest advances are in improvements to the way users utilize Google Search on Android. Now users can take advantage of Google’s Knowledge Graph along with voice search to get instant spoken answers via Google. So users can ask specific questions with voice search and Google will spit back specific answers on an individual “card.” Users can also access the same information using a regular text search. That gives users direct answers, kind of like Siri, but it also allows people to access regular Google search results with one swipe.

Google is also introducing Google Now, a new way to search that is like a contextual personal assistant. From the Google Search page or with a swipe up from the bottom of the screen, users can access Google Now. The idea is to give people instant access to information based on a person’s location, time of day and search history. So if Google knows you commute to work every day at the same time, it will provide a card showing how long it will take to get to work, offering faster alternatives if the usual route is slow. For appointments, it will tell you how long it takes to get there on time and will show options to get there. If you’re catching a flight, it will tell you if the plane is delayed, what terminal it’s in, and when to leave.

Google Now will also look at a person’s location and show interesting restaurants and places to visit nearby. Users can peruse a restaurant’s menu or check out the best things to eat there. When a user is at a bus stop or transit station, it can show when the next bus or train is coming. The location feature can also tell when a user is traveling and can surface cards with information on currency exchange rates and the local time at home. Sports fans can see sports scores for their favorite teams. This can lessen the need for users to consult other apps such as Yelp, Foursquare, ESPN and other travel and transit apps.

Google’s voice-recognition service for text input has been enabled for offline use so users can dictate messages without having to be connected. The text-entry system has also been improved to include predictive typing. There’s also automatic widget resizing when you move one around, so a widget can shrink when you relocate it to a crowded page.

Android notifications are getting a lot more powerful. Users will be able to call back people, see emails, control music apps, email upcoming meeting participants, and do other actions right in the notifications pane. Each notification can be enabled with up to three actions. Notifications can also be resized with a two-finger gesture.

Performance-wise, Jelly Bean provides some subtle performance upgrades such as improved system frame rate, triple buffering and making everything work in lock-step to vsync, which runs across all the drawings and animations. That makes rendering smoother and quicker.

Google also introduced Google Cloud Messaging for Android, a service that lets developers send short message data to their users on Android devices. The service, which doesn’t require any special syncing, will order and deliver messages, with up to 1,000 devices able to be reached with a single request.

Other improvements include the ability to transfer videos and photos among devices using Android Beam and Bluetooth and better accessibility for gestures with more tutorials and bidirectional text support for languages that move right to left.

Here’s a video showing how Google Now works:

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Analyzing the wearable computing market
• The fourth quarter of 2012 in mobile
• The 2013 task management tools market