Negative Approach

Looks like Google is starting to take Apps Premier a bit more seriously, providing users with SLA credits for the August outage and realizing that the guy who chose Google Apps has to report to someone else when things go haywire.

Not much more to comment on...I hope they make it work. I switched entirely to browser-based Gmail and I'm still liking it.

Email pasted below from the Google Apps team:

Given the production incidents that occurred in August, we'll be extending the full SLA credit to all Google Apps Premier customers for the month of August, which represents a 15-day extension of your service. SLA credits will be applied to the new service term for accounts with a renewal order pending. This credit will be applied to your account automatically so there's no action needed on your part.

We've also heard your guidance around the need for better communication when outages occur. Here are three things that we're doing to make things better:

1. We're building a dashboard to provide you with system status information. This dashboard, which we aim to make available in a few months, will enable us to share the following information during an outage:

1. A description of the problem, with emphasis on user impact. Our belief is during the course of an outage, we should be singularly focused on solving the problem. Solving production problems involves an investigative process that's iterative. Until the problem is solved, we don't have accurate information around root cause, much less corrective action, that will be particularly useful to you. Given this practical reality, we believe that informing you that a problem exists and assuring you that we're working on resolving it is the useful thing to do.

2. A continuously updated estimated time-to-resolution. Many of you have told us that it's important to let you know when the problem will be solved. Once again, the answer is not always immediately known. In this case, we'll provide regular updates to you as we progress through the troubleshooting process.

2. In cases where your business requires more detailed information, we'll provide a formal incident report within 48 hours of problem resolution. This incident report will contain the following information:

a. business description of the problem, with emphasis on user impact;
b. technical description of the problem, with emphasis on root cause;
c. actions taken to solve the problem;
d. actions taken or to be taken to prevent recurrence of the problem; and
e. time line of the outage.

3. In cases where your business requires an in-depth dialogue about the outage, we'll support your internal communication process through participation in post-mortem calls with you and your management team.

Disclaimer: The opinions represented here are my own and do not necessarily reflect those of my employer.

From Cloudsourcing to Cloudstorming to Cloudburst (which can be both negative and positive) a new set of terminology has arrived. It was inevitable. Embrace the Cloud.

Over on Thinking Out Cloud, Geva Perry outlines some of the new Cloud Computing Terminology, which does start to sound a bit like Dr. Seuss after a while.

These new terms help to explain the various aspects of the Cloud and somewhere down the line it starts to show that the internet and SOA have come together. Nonetheless, make sure you hold your breath lest you drown in the metaphorical sky.

Disclaimer: The opinions represented here are my own and do not necessarily reflect those of my employer.

My post on using the Cloud for storage went live just minutes before my intrepid IT guy Kevin received this email from utility computing provider Flexiscale about the potential large-scale loss of data stored on their Cloud storage service.

The short version: human error in their backup process deleted one of the main storage volumes. Roughly 12 hours later users have read-only access to the storage platform but no read-write. And now, they have to rebuild, but don't have the space.

"After consulting with our storage vendor it was agreed the most sensible option would be to copy the entire volume to a new disk structure (still maintaining it's integrity and structure), from where we could re-mount it correctly. Unfortunately due to it's size we didn't have spare capacity on the platform to create a complete duplicate of it."

Without disparaging Flexiscale, this is what I mean about the BigCos like IBM figuring these "enterprise-class" features out before enterprises move into Cloud consumption.

Full email pasted below:

The vast scale of services like Amazon S3 or Google Apps provide new ways to establish or augment backup and disaster recovery plans.

Realistically, many large corporations won't trust their data to Google or Amazon.com (and they probably shouldn't). However, they will trust IBM, AT&T, and other big companies, as they start to expand their offerings.

If you are using AT&T bandwidth, and you can tap into its data centers for data recovery, you've just solved a major problem in a theoretically secure manner.

Managed backup and data recovery services do exist today, but they tend to be very expensive "enterprise-class" or very mediocre consumer-oriented services. There should be a way for cloud infrastructure to become a real option for enterprises.

Here are a few of the issues that would need to be addressed:

• Automation: How does the data get from internal servers to the cloud, or how does it get from individual desktops?
• Security: What is the security model that can be applied and managed universally?
• Data integrity: How do I know that my data is actually my data if I am not in private space or virtual machines?
• Risk: What is the risk of losing my data without a defined service-level agreement?

The broad adoption of virtualization can be very beneficial but can also present some challenges, considering the lack of tools and options available. And considering that virtual machines are what most of today's clouds are built on, odds are that you will run into the very real challenge of how you even manage your virtual machines effectively.

According to the latest disaster recovery research report from Symantec (NASDAQ: SYMC), based on surveys of 1,000 IT managers in large organizations worldwide, 35 percent of an organization's virtual servers are not included in its disaster recovery plans.

Worse yet, not all virtual servers included in an organization's disaster recovery plan will be backed up. Only 37 percent of respondents to the survey said they back up more than 90 percent of their virtual systems.

One way this lack of attention to disaster recovery could be addressed would be direct connectivity from one cloud-computing provider to another, giving users the ability to replicate data. Alternatively, users could set up virtual-machine farms internally that push the data to the cloud as a backup.

It won't be long before we start seeing storage clouds popping up. My bet is on the big IT vendors to get there first.

The more research I do into online video games, the more I see that there are still very large opportunities available for those who know how to develop games and build community. When the market leader in a category has only 11.5 percent of the market that says that there is plenty of room for other challengers to take pieces of the pie.

The Hitwise data listed below for July shows an interesting split between aggregator sites like Pogo and Yahoo and specialized game sites like RuneScape and Webkinz showing that you can make money (or at least get eyeballs) via either method of gaming.

Note: The Hitwise data featured is based on U.S. market share of visits as defined by the IAB, which is the percentage of online traffic to the domain or category, from the Hitwise sample of 10 million U.S. Internet users.

Better get a look at this awesome stop motion video of a bunch of Legos performing Whiplash, a Metallica classic, before Lars Ulrich sees it and sues everyone.

BTW--I heard the new Metallica song on the radio and it sounds like they just chopped up 4-5 old songs.

Via GeekStuff

If you're a Java developer and use Ant or Maven to build your software, there's news coming out this morning that should catch your attention. Sonatype, the company built around supporting Maven, is announcing its Nexus 1.0 release, a repository manager that allows developers and teams to quickly and easily manage internal and external repositories, including the Maven Central Repository.

Behind a build and release framework, managing big repositories of artifacts and tools adds a level of complexity to software development that can slow or kill a build. Great idea, having a big bucket of stuff to pull out code snippets and tools whenever you need, but not simple. Nexus 1.0 addresses this directly.

Jason van Zyl, creator of Maven, the popular Java development framework, has quietly founded an open-source company around the popular build framework. Maven has over 2 million downloads and the Maven Central Repository--home, incidentally, to more than 50,000 main artifacts and averaging 100 million hits a month--has made access to and integration of Maven crucial to Java developers worldwide.

Sonatype has done a couple press releases this summer about joining the Eclipse Foundation and releasing a plug-in that integrates the two. But otherwise they've remained quiet about their audacious goal. Maven plus Eclipse going after Visual Studio plus .Net. This is a serious and popular open-source alternative to Microsoft's development juggernaut.

This is an open-source company worth keeping an eye on.

BONUS: O'Reilly and Sonatype have published a much-needed reference book, "Maven: The Definitive Guide," covered by a Creative Commons license and replete with examples. It's free online in html or PDF or can you can order it through O'Reilly.

Disclosure: My company shares investors with Sonatype.

Cloud computing means different things to different people. In this guest post, Tom Mornini, CTO, of Engine Yard looks at the differences amongst the applications that get lumped together as one amorphous cloudy mass.

Navigating the cloud
Guest post by Tom Mornini, CTO, Engine Yard

The term cloud computing is now in common use. So common in fact, that it clearly encompasses many incompatible ideas. Let's consider the differences among Amazon's AWS, Google's AppEngine, and Apple's MobileMe.

AWS provides "raw" compute resources via EC2 and higher level services such as S3, SQS, and SimpleDB. You configure the raw compute resources as you see fit, using the OS of your choice and write applications in any language using any framework. Those applications may or may not rely upon additional AWS services such as S3, SQS, and SimpleDB. Deployment is entirely up to you.

AppEngine, on the other hand, is far more opinionated. Just as Ford once sold cars of any color, so long as they were black, AppEngine allows you to write programs any way you want, so long as it's their Framwork, Python, and BigTable. While Google has already suggested that it would support other languages in the future, it seems clear that Google doesn't intend to offer as many "colors" as AWS does.

This is not to suggest that AWS is "better" than AppEngine or MobileMe, merely different. As Engine Yard is 100 percent committed to Ruby and Rails, we're big believers in the Rails creator's triple philosophies of: "Convention over Configuration," "Flexibility is Overrated," and "Constraints are Liberating." It certainly does mean that AppEngine is dead to Engine Yard, since we simply cannot make use of it using the technologies we prefer.

Apple's MobileMe marketing also makes reference to the concept of cloud computing, but takes it in an entirely different direction than Amazon and Google. It doesn't offer any software development or deployment at all.

From my perspective, EC2 is a hardware as a service (HaaS) offering while the rest of AWS is a platform as a service (PaaS) offering. AppEngine is a Platform as a Service (PaaS) offering that also provides deployment as a service (DaaS). MobileMe is a software as a service (SaaS).

If cloud computing encompasses all of these items, it seems to me that we've developed a new name for something we're all quite familiar with: the Internet! Consider this: When, exactly did subscription Web sites become known as SaaS? I suspect that this term is rooted in enterprise marketing. After all, nobody would ever pay that much to use a Web site!

Programs running on computers have been providing flexible and valuable services to the cloud for quite some time now. Since the beginning most end users have been unable to conceptualize how it works. Sun deserves credit here: "The network is the computer" was really ahead of its time.

So, perhaps cloud computing represents a new abstraction layer entirely. This time around even the developers don't understand how their applications work. Of course, many might suggest that this has always been the case!

As applications grow more complex and become interconnected, simply managing applications becomes a highly complex task. Particularly so as consumers of these applications demand local application response times and 100 percent uptime. These consumers don't care and don't want to know about peak loads, hardware failures, etc. The providers of these applications are increasingly finding little benefit in managing the infrastructure themselves.

For the cloud to reach its full potential, I strongly suspect that new application development technologies will be required. With dynamic hardware availability, applications must be dynamic. Security, resiliency, and scalability must be inviolate. Most importantly, all of these ideals, currently recognized and strived for by developers, must be inexpensive to achieve and built into the very fabric of these technologies.

It's imperative that these technologies provide enough abstraction to allow an application to run not just on a single cloud, but be run and managed across the one true cloud: The Internet itself.

Tom Mornini, CTO
Engine Yard.

With few reference points for enterprise cloud computing consumption, many new and interesting companies are popping up to make cloud resources available, with the aim of simplifying the processes.

3Tera's CEO Barry X. Lynn wrote a guest post on ZDNet that got me wondering about how software companies can leverage tools that make the cloud more consumable for users and, specifically, enterprises. Lynn takes the view that operations will be abstracted completely from data in the future, which affects both enterprises and the software that they run their businesses on.

While enterprises are growing comfortable with applications in clouds and realizing the upside of dynamic provisioning and scaling, they will be developing new applications and replacing/changing existing ones. They will start building the new applications in clouds and as they change existing applications, will consider migrating them to the cloud in the process. This will afford them the advantages of much faster time to market, the ability to run applications on demand in multiple data centers (globally if appropriate) creating their first truly complete disaster recovery abilities and concentrate on their core businesses which may be financial services, health care, manufacturing, etc., but certainly is not data center operations. (They will leave that to the companies whose core business IS data center operations.)

I spoke to Barry about how 3Tera works and got a demo of the service. My overall impression is that this is a very powerful tool set that is way ahead of how people are utilizing cloud-based resources today. I'm just not sure that the approach is the right one for the masses.

3Tera's Applogic allows you to abstract the hardware from the software and adds a management layer. Anything that can run in a physical data center can run in an Applogic cloud without having to change any code.

Applogic encapsulates all the virtual resources as "applications"--OS, apps, servers, etc., and let's you superimpose the combined components as an image on a cloud or grid. Applogic will then consume the appropriate the resources from the underlying hardware and software.

Applogic provides a browser-based Visio-like UI that lets you create these "applications" by dragging in resources that create a definition file that describe the overall package. You don't need to know anything about the infrastructure that an application or service is running on; you just need to define the level of resource availability and service level required.

All of that makes a lot of sense, providing a virtualized infrastructure with the software and hardware running separately but creating a complete environment.

The question in my mind is if this "virtualized layer" is what enterprises want. The other question is if cloud consumers will be interested in this type of Visio-like approach or will they just want to "mount" cloud resources that are managed from behind the firewall.

Of course, the biggest question is if the cloud is anything more than "mainframe in a browser."

Ready to battle Iron Chef Morimoto?

(Credit: Destineer)

Iron Chef America: Supreme Cuisine lets players square off in Kitchen Stadium and battle through a series of fast-paced and intense culinary challenges. Each victory advances players closer to a final showdown that will determine who will reign supreme as the next Iron Chef America.

I hope the bonus rounds let you battle it out with the real Iron Chefs. Give me Hiroyuki Sakai and Rokusaburo Michiba any day over these Iron Chef America guys. And don't forget floor reporter Shinichiro Ohta shouting out "Fukui-san!" every few minutes.

I've written in the past about why the Wii is the ideal platform for video games that have more interaction than those console games that typically only require you to work out your thumbs.

And lest you think that I am all about Wii violence, I like the idea of cooking that lets all us Food Network junkies live out the chef fantasy without having to do any real work.

As a side note I ate at Iron Chef Chen Kenichi's restaurant in Akasaka about five years ago and the food was unrecognizable (despite claiming to be Chinese), but just think of all the fun gameplay that entails.

Via EaterSF