Tom Lounibos | SOASTA

What Rackspace’s OpenStack Announcement Means to SOASTA Customers

Posted by Tom Lounibos on Apr 18, 2012. Leave a Comment

In case you weren’t able to attend Monday’s Rackspace OpenStack Design Summit in San Francisco, here are the highlights:

On the heels of rival Citrix Systems’ CloudStack announcement, Rackspace released its production-ready next generation cloud powered by OpenStack. OpenStack is the open source cloud operating system defined by developers and organizations including NASA, IBM, Red Hat, Morphlabs and Piston Cloud Computing.

What does this mean for SOASTA customers?

We grew up with Rackspace in the Cloud. Since 2008, we’ve used tens of thousands of Rackspace servers. In fact, this past weekend we ran a millions-of-users test for the 2012 Olympic Summer Games. We used Rackspace Cloud Servers to help Verizon test in preparation for an iPhone launch. We used it to help Hallmark prepare for Valentine’s Day.

We’ll support the OpenStack Cloud Services by offering the Rackspace Cloud as one of the public cloud choices available for the SOASTA Global Test Cloud. Our Global Test Cloud, part of our CloudTest Platform, provides our customers with the ability to simulate hundreds to thousands to millions of users across the globe to ensure the production performance of their mobile and web apps.

Like Maverick and Goose in “Top Gun,” we feel the need for speed. The quicker we are, the more tests we can run. For us, speed and capacity go hand-in-hand. On a daily basis we’re accessing thousands of servers for testing applications – often in production – across data centers, geographies and service providers. At last count our Global Test Cloud was made up of 350,000 servers from 17 service providers in 47 different locations. We run a lot of tests and we need a lot of capacity. And we’re always looking for more.

We look forward to our continued partnership with Rackspace.

Evolution vs. Revolution: What will it take for testing join the 21st Century?

Posted by Tom Lounibos on Apr 04, 2012. 4 Comments

Put aside, for a moment, the extraordinary events of the last 12 months. Then hear me out on this one, which falls into the category “It Generally Goes Like This.”

While great ideas ignite the flame for change, real change, meaningful change comes from the advancement of an original thought.

Here’s an example:
Henry Ford did not invent the automobile. It’s widely assumed that it was Ferdinand Verbiest, a Flemish member of a Jesuit mission in China around 1672. While Verbiest was clearly visionary, the world wasn’t ready to embrace the idea until 1906 when Henry Ford made cars affordable by optimizing their production. This changed the world. Although the rate of change has advanced quite a bit since the 17th Century, it still takes a few years to evolve even in this, The Digital Age.

Now let’s relate this to our industry. There has been some progress in test automation over time. However, significant change, change along the lines of what we have seen in Operations (deployment) with cloud computing, has been elusive for the testing business. At some point you have to ask why. Roughly 80% of all mobile applications today are still manually tested. That’s somewhere around 600,000 apps!

Surely I am not the only one who thinks this begs for improvement since so many companies are reducing the cycle of development to one or multiple times per day. The impact of this slow adoption of test automation is felt every day by consumers like you and me as we impatiently wait for apps to be loaded. It’s irritating when our airline cannot process our reservation or when we have Bruce Springsteen tickets in our shopping cart, only to see that they disappear before we enter our credit card number.

Innovation has led us into the Test Automation Renaissance that we’re currently enjoying. There are new vendors and new technologies making this easier and less expensive. But why, then, is so much testing still being done manually?

It’s Springtime For Mobile Application Testing

Posted by Tom Lounibos on Feb 09, 2012. Leave a Comment

In the spring of 2008 Cloud Computing was the featured “flower” that dominated the conversation in the technology flowerbed. With Spring 2010 arrived a new flower, Dev/Ops, with its Agile methods, practices, and integration between software development and IT operations. Now, in the spring of 2012, another flower is blooming: Mobile Testing.

Testing, once called the bastard child of IT, is having a significant renaissance brought about by the emergence of a global consumption model and the subsequent mobile application explosion. With customer service and now revenue heavily influenced by application performance, the past three years have seen a significant influx of innovative test approaches arriving in the mobile development market. And all this in hopes making application performance a differentiator for their consumers.

Of these new approaches, cloud-based tools have received the most attention. In a recent IDC survey of 350 of the leading consumer-facing websites, 17% of respondents reported that they have already begun to use cloud-based test solutions. Even more remarkable is that 43% of respondents indicated that they are planning to use cloud-based solutions in the next 24 months. These are promising growth numbers for any emerging market and they are especially promising for this market.

Why is Cloud Testing blooming now, after vendor-introduced solutions have been around for more than three years? IDC’s Melinda Ballou, who conducted the survey, says, “Cloud testing is a hyper-growth market because it just makes sense”. Mobile applications are now being used by thousands if not millions of consumers.

Old, process-based approaches to application testing cannot keep pace and are too expensive for this new breed of mobile developer. Gartner’s Tom Murphy says that many will rely on manual tests and a prayer to overcome the challenges of gestures, geolocation, motion and realistically conducting load tests.

The speed, agility, scale and cost savings of test solutions built specifically to leverage the cloud make them especially compelling in the mobile application development community.

Another sign of this early spring has been an upsurge in acquisitions of test technologies during the past six months: ITKO (CA), GreenHat (IBM), and Blaze (Akamai). For all of these reasons and more, spring has come early this year for Mobile Testing and it appears that we can look forward to seeing this particular flower for many years to come.

“Minority Report” Defining Today’s User Interface

Posted by Tom Lounibos on Jan 18, 2012. Leave a Comment

Think back to Steven Spielberg’s 2002 classic, “Minority Report”, the movie that explored the future of human/machine interface in 2054. It was both cool and exciting, and it was also years away from reality. Or was it? Here we are, now ten years removed, and some of these concepts, gesture-based applications (everywhere), hovering gestures (Apple patent), 3D navigation (everywhere), and voice (Siri), are very much real.

John Underkoffler is the UI designer who built the technology for Minority Report and in real life, too. His 15-minute TED presentation is well worth watching.

From logistics and supply chain to financial services, new applications evolve every single day to leverage advancements in man and machine user interfaces. In many cases there is simply no other option than to rapidly wade through the enormous amounts of data these applications generate in search of the actionable intelligence.

One of these Big Data issues occurs while performance testing web and mobile applications. These applications simulate millions of people around the globe hitting a web site or mobile application to provide predictive analysis as to where performance bottlenecks may occur. In doing so, they generate terabytes of performance-related, real-time data. Analysis of this data is extremely difficult using traditional, one-dimensional test tools because problems are often buried amongst layers and layers of data.

Over the past few years, companies such as Splunk and SOASTA have emerged because their multi-dimensional capabilities enable testers to see several pieces of performance data on the same timeline and in a single view. These kinds of user interfaces allow for real-time resolution to problems such as performance for a growing number of leading companies.

Of course new problems arise as these problems are solved. The next generation will surely struggle with delivery of a stable user experience when application UIs are gesture-based, voice-based or even 3D. I will leave that question for another day. The bottom line is this: Minority Report has arrived. Is your team ready?

Open or Closed? Technology’s version of The Beatles vs. The Stones

Posted by Tom Lounibos on Dec 20, 2011. 1 Comment

Your preference for one band over the other speaks volumes about your personality. Do you prefer the band with blues in its DNA or the one with more pop standards in its repertoire? This debate was enough to divide men born only a few years apart.

A similar debate has raged on in the Silicon Valley since 1977: Open vs. Closed Architecture. While there have been periods in technology time where the issue was less pressing, today the matter is white hot again as new software platforms emerge around cloud computing. In fact, while the decision to go one direction versus the other is not as impactful today as it was in 1977, the issue has influenced many products since. And we’ve seen that the two philosophies can breed products that co-exist in some form of harmony, or at least in market opportunity.

Back in the 70s, the most public debate over this came in the form of hardware by the two founders of Apple. Steve Wozniak and the Homebrew Computer Club believed in open architectures (ports in those days) that would allow devices to be added easily by other engineers. Steve Jobs, meanwhile, was all about the control and optimization that he felt could only be achieved through closed architecture. Wozniak won the first battle and the Apple 1 computer had four ports. Yet Jobs won the war and built the company around the Macintosh, which had virtually no ports. And the rest is Silicon Valley history.

Today the battle continues between Linux and Windows, OLTP and closed transaction management systems (e.g. IBM CICS), the Internet and private networks, and Android and iOS, among others. And now it continues into the Cloud Platform wars.

The open source community is known for inspiring innovation and causing previously monolithic applications to become commoditized. Today, open source cloud platforms like Cloud Foundry, Cloud.com, OpenStack and Eucalyptus offer the promise of do-it-yourself clouds in a secure, private test lab before moving to either private cloud or public cloud.

Dan Kusnetzky recently wrote this thought-provoking blog post, Grocery vs. Restaurant, OpenStack vs. Amazon, which compares open and closed architectures to grocery stores and restaurants. An open source model is the grocery store, which provides users or shoppers the tools or ingredients to assemble their own meal. Closed architectures are restaurants: someone else selected the tools or ingredients and has packaged them into a meal.

All of this history begs the question: which is better? If history provides any indication, we’ll still be debating this issue for the unforeseeable future. Where do you weigh in?

Growth or Liquidity: The Funding Question of 2012

Posted by Tom Lounibos on Dec 11, 2011. Leave a Comment

On Friday, a little-known e-commerce payments company named Klarna raised a massive $155 million round of funding from DST Global and General Atlantic. Its previous round was only $9 million, in May 2010, when it was discovered by Sequoia Capital and superstar partner Michael Moritz took a board seat. Of this investment Moritz says, “This is the public financing of twelve years ago. It is just done privately.”

Today, founders and board members of cloud and mobile startups have new questions to consider when raising capital. For example, do the founders need or want to take some liquidity at the same time they take additional capital? For a few lucky entrepreneurs like Klarna, Facebook and Zynga, this new funding strategy has been a brilliant way to bridge the gap on the long road toward an IPO or acquisition. The cynics see it as just another control strategy used by venture capitalists. Either way it is clever.

The real question that this funding strategy poses to an entrepreneur is: Just how confident are you in your idea and in your team? If you have some doubts or, frankly, you need the personal capital, then an early liquidation strategy is a very effective remedy. But, if you’re absolutely certain of the game-changing nature and unique value that your solution delivers, why take early liquidity unless you have to? For the many young entrepreneurs who have put themselves through college and then struggled to start their company by taking little or no income, taking this money and running makes a lot of sense. Yet for others that may have enjoyed some success and who are most confident in their offering, the better strategy remains to pull out as little capital as needed to become the winner in your space.

That said, my company, SOASTA, is pleased to announce our newest funding of $12M. To date we’ve raised $32M from Canaan Partners, Formative Ventures, Pelion Ventures and TEFIII. We would like to thank our thousands of customers, hundreds of employees and partners around the world. Season’s greetings and best wishes for 2012.

The Dawn of a New Shopping Season

Posted by Tom Lounibos on Nov 29, 2011. Leave a Comment

As Thanksgiving Eve, Black Friday and Cyber Monday approach, the momentum is palpable as eager shoppers gear up to take advantage of all those well-advertised holidays sales. For many retailers, as much as 60% of their annual revenues are dependent on the next six weeks. No mistakes are permissible; inventories have to be stocked, clerks must be ready to assist everyone and anyone, and most importantly, the technology supporting sales operations must work.

In fact, technology is at the core of a successful holiday shopping season as a growing number of consumers go online, stressing applications, infrastructure and networks. The most dedicated shoppers began shopping online in earnest before Thanksgiving — to the tune of $9.7 billion according to a report released last week by comScore.

Faulty technology can cause retailers to lose millions of seasonal dollars. If a retailer’s web site fails, shoppers pursue alternatives. And today there are more retail options then ever before, those options being only two clicks away. So, retail friends, if your site is not easy to use and the buying experience is not simple and fast, it’s likely that you’re losing sales.

Many retailers spent this year making over their web sites and are using some of the coolest technologies to highlight their holiday season product offerings. As hip as these features are, they could, in fact, crash a site or deliver a not-so-positive user experience. This year, at least six of the top ten retailers have prepared their web properties for the online shopping season.

These savvy consumer brands spent months testing these new technologies and apps. They tested and re-tested new functionality on every consumer-facing application hundreds of times. They also began to embrace cloud testing techniques to simulate millions of customers buying leading brands via iPads, iPhones and Android. And they did this all to ensure that their user experience is easy and fast.

It’s has been a long summer for most retail development and test teams. Showtime is here! Good luck to the retail community; we wish you a prosperous holiday season. And, to all you consumers: enjoy the new sites and the high performance that they will deliver.

The Day Apple Lost its Cool

Posted by Tom Lounibos on Nov 22, 2011. Leave a Comment

Last week I concluded our company meeting by playing Apple’s 1997 Here’s to the Crazy Ones commercial. I chose not to use the original version narrated by Richard Dreyfuss and instead went with the version that Steve Jobs himself narrated. It has always inspired me and I wanted to inspire our team. It also highlights what I think was Steve’s greatest quality: his passion for what he truly believed in. This gave him the ability to inspire millions. When the commercial finished playing, and even though I had seen it a hundred times, I had a tear in my eye and I struggled through the meeting’s closing comments. You see, Steve Jobs evokes emotion from almost everyone he has touched.

Truthfully, back in the late 70’s and 80’s I was not a huge Steve Jobs fan even though we had both grown up in Silicon Valley tech world just a few months apart. Back then we didn’t need LinkedIn as you were only two or three degrees of separation from most everyone else in the Valley. Steve and I shared mutual friends and common business relationships, and we even competed against each other over object oriented programming languages when I was at Digitalk and he at NeXT. Truth be known, I turned down a job offer to work with Steve at Apple in 1982, a decision that I now regret.

So why wasn’t I a fan? If Steve Jobs was your professional barometer it was hard feel very successful. That is, if you were ever stupid enough to try to match up with his success. He had founded and built a $2B tech company before I even knew what I wanted to do with my life. Simply stated: I was jealous!

That all changed for me in spring of 1996 when Steve headed back to Apple. I love people who overcome adversity and there has been no one in the Silicon Valley who has been knocked down harder then Steve Jobs was back in 1985. It is a well-documented story of rebirth. From the moment that NeXT was bought by Apple I found myself rooting for him to win back the company that had ousted him 11 years prior. This, even though just a year before he had famously described our merger of the two Smalltalk companies (Digitalk and ParcPlace) this way: In a glass of water, two rocks will sink as fast as one. Then came his Crazy Ones commercial, which soon became the informal Declaration of Independence for all of us Silicon Valley entrepreneurs. Steve Jobs was once again our symbolic leader. Steve was cool again.

We have lost one of the great crazies of the Silicon Valley, one whose passion and spirit inspired millions to “think different”. Apple cannot replace Steve Jobs but he left the company with an enduring legacy of cool, one that I hope they never lose again. RIP Steve. And thank you.

The Future of Testing: Evolve or Elimination. It’s a Choice.

Posted by Tom Lounibos on Nov 17, 2011. 1 Comment

I’ve been seeing a lot of tweets and blogging around the future role of the Software Tester. Some have suggested that, with all the recent advancements made in test automation, the test community may face employment challenges. I strongly disagree. That said, rapid technology innovation necessitates that users adapt and evolve, and the test community will not be immune to this requirement.

The fact is that today, in 2011, the majority of Testers are still relying on the same tools that they used in 1989. How is it that this community has been not been forced to change all that much over the past 20 years? This is highly unusual and counterintuitive in an industry that has long been affected by term limits for technology, one that is known for discarding old tools in favor of new tools to support emerging technologies.

How is it that so many Testers still use tools borne during the Reagan Era and when client server applications dominated the tech landscape? Have our applications neglected to evolve? Of course not!

One theory is that Testers are just wired differently, that they are more resistant to the forces of change. If this is the case, consumers beware: the impact will surely lead to an increase in web and mobile app failures. Why? If Testers hold onto previous testing practices then today’s dynamic applications will need more and more frequent performance testing because of the demands for speed, scale or cost savings.

So what needs to happen?

Like other professionals, Testers must evolve their skill sets. They must acknowledge and respond to the pressure to deliver actionable intelligence back to the business as a result of their testing. Further, they must begin to understand and accept responsibility for all aspects of delivering performance. Performance is the correlation of Application, Infrastructure, Network and Users, and a valuable, savvy Tester will deliver aggregated and correlated results to support their actionable intelligence.

It’s a new world of HTML, gesture-based UOs, Cloud, streaming video and voice, and Testers must embrace the opportunity or they will find themselves out of work. Carpe Diem Testing. Adapt and Evolve. Join me in helping the testing community become part of the cognoscenti, be on the forefront of this technology adoption lifecycle, earn industry respect by embracing testing’s place in the revenue cycle, and be the leaders in delivering the next generation of High Performance Applications.

Mobile Apps Will Require Application Performance Management 2.0

Posted by Tom Lounibos on Nov 09, 2011. Leave a Comment

Application Performance Management (APM) 1.0 was defined years ago by Gartner as application performance monitoring for client server applications. Players such as CA, Wily, IBM, HP and BMC dominated the landscape, which was characterized by delivering real-time “status” of how a customer’s infrastructure was running in production. Then, as new web applications started appearing around 2001, a new set of players began to emerge. These included Gomez and Keynote, which added monitors around the end-user experience. Then recently, another breed of APM vendors came along focusing on monitoring customer transactions. These vendors include OpTier, AppDynamics and New Relic. In general, companies have two or three vendors’ monitors tracking their production status. Is this true management?

Today’s customers are looking for the next generation of Application Performance Management tools, tools specifically built for mobile applications and which track all aspects of performance. We’ll call this Application Performance Management 2.0.

There are two components of APM 2.0. The first is the delivery of a single real-time view of all performance metrics on the same timeline. This enables companies to drill into performance data much deeper/faster because it’s all located in the same place thus giving these companies, in effect, a multi-dimensional view of their performance data. Why is this so? Performance issues occur from a cascade of events eventually leading to latency or even a crash. Multi-dimensional views get to the problems quicker.

The second component of APM 2.0 is the need to add simulation to their platform. Today, companies can only react to a fire drill. Instead, companies want to prevent this fire drill from occurring in the first place. To do this requires the ability to simulate hundreds of different scenarios, which include massive loads, unique devices and many other variables.

Preventive maintenance through simulation will be the key to 2012 APM 2.0 tools for almost every mobile app builder.

Older Entries»

Archives

Recent Posts

Recent Comments