Tag Archive for ‘cloud computing’ rss

Agility. At what cost?

Posted by Tom Lounibos on Nov 03, 2011. Leave a Comment

For the past several years those of us in tech have been in a race to reach a growing global marketplace of consumers newly armed with mobile devices.

Mobile apps are different from the previous generation of apps in how they are built (assembled), by the type of dynamic content they contain (video, etc.) and by the speed in which change occurs. So much so that a new buzzword now defines their nature: Agile. In this instance Agile means the ability to refresh an application a couple times a day. To do this, most companies have had to rethink their current development processes which, in many cases, were structured around monthly release schedules.

As the race to meet the growing mobile market ramped up, new platforms such as cloud computing emerged to help deliver agility in the deployment process. Then new development languages (Ruby) and technologies (Ajax, HTML5) arrived, adding agility to the development cycle. The race was on, and business agility became more and more of a daily reality. The problem: At what price was agility being achieved?

Behind the scenes, developers began to believe that their new mobile apps would never fail if they had access to unlimited and affordable compute power to run them, as was promised by cloud computing. So they stopped testing these apps in a rush to bring new features live. For many, this single shift of eliminating the test process has destroyed all the good that the advancements in cloud computing and new Agile development technologies had delivered.

In ecommerce, though, a website has to work because the consumer is only couple clicks away from the competition. When these sites began to fail, consumers went elsewhere and they may never come back.

Great progress has been made in delivering corporate agility. However, as important as agility is, performance trumps agility on consumer-facing web sites every time.

 

What Keeps a 100-Year-Old Company Feeling Like a Teenager?

Posted by Tom Lounibos on Jun 06, 2011. Leave a Comment

This year IBM turns 100 years old, and last week IBM’s “Watson” was named 2011 “Person of the Year” by the Webby Awards. The mere thought of IBM as a start-up (circa 1911) boggles my mind, especially when you consider that this year they reached their all-time high water mark at $205B in market valuation. This feat is even more amazing given the somewhat rocky road that they traveled during the late 80’s and 90’s. Over the last ten years under the watchful eye of Sam Palisamo, IBM is beginning to experience a rebirth. While they are nowhere near the dominant leadership position that they held over the technology sector from 1940 – 1980 when the market was defined as “IBM and the Seven Dwarfs,” today they are beginning to show signs of re-emergence as — if nothing else — the “supervising adult ” that their 100 years of existence entitles them to.

So what keeps an old company relevant after all these (100) years? Probably the same things that keep older people young…they hang out with their kids or grandchildren. In IBM’s case they have learned (over the years) to partner with a few young, innovative companies that are proving to be the new “game changers” in several traditional markets — even in some markets that have long been considered to be IBM’s strongholds. By partnering up with these young “upstarts,” IBM has given their customers fresh alternatives for new technologies and approaches for dealing with a rapidly changing business world. Perhaps even more importantly for many of their customers, IBM is also delivering a much-needed layer of “adult supervision” in this increasingly crowded and complex vendor landscape. Their years of experience enable IBM to become a trusted advisor to their customers on how to navigate through this vendor mine field.

As one those lucky few young upstarts that Grandpa IBM has chosen to partner with, we here at SOASTA get the advantage of their many years of experience surviving and thriving in both up and down markets. We also get to benefit from what may be the greatest technology distribution channel ever compiled, a channel that no start-up could ever replicate organically.

Time will only tell if Grandpa IBM and its young upstart partner SOASTA will make an interesting combination for the IBM nation, but the hourglass has been turned. One thing is for sure, SOASTA is one youngster that is eager to learn from Grandpa.

Will the Cloud Change the Industry Giants or Will the Giants Change the Cloud?

Posted by Tom Lounibos on Jun 01, 2011. Leave a Comment

Over the past couple of years or so the technology giants such as IBM, HP, Cisco, Oracle and Microsoft have all announced cloud strategies — all with promises of multi-billion dollar investments backing these strategies up. Some have even predicted that cloud computing is the key to their future over the next ten years. This type of affirmation from these industry leaders is pretty heady stuff for a cloud industry in its own infancy…but will it become a reality? Will the industry leaders of today’s technology market be the leaders of tomorrow’s?

Most people following this market will say “absolutely,” even though the majority of cloud innovation continues to flow from early stage start-ups in this space today. It is also clear that the technology giants have both the cash reserves and the desire to take a leadership position in the cloud through acquisition. But is this enough?

The biggest obstacle to the giants achieving their ultimate goal of owning the cloud may be in their existing business models. The very same business model that has been the envy of the industry for the past thirty years may, in fact, be a roadblock to owning the cloud unless something changes.

Companies such as IBM, HP, and Oracle have been built largely on top of enormous distribution channels of direct sales and support personnel to attract new and maintain existing customers. While these channels have been the envy of the industry for the past thirty years, they also come at an enormous cost that must be compensated for in these companies’ pricing models. Conversely, the cloud’s on-demand pay-per-use business model (often defined as the “anti-lock-in” model) may be the antithesis of the enterprise site license that was popular in the 80’s and 90’s. Most importantly for publicly traded companies is that the cloud’s on-demand model offers limited visibility into projected or future revenue streams, which is a major issue in delivering guidance to share holders. Even worse, the average deal size for on-demand servers ranging from $.10 to $.80 per server hour rarely, if ever, goes over $100,000, let alone the $1M threshold that commonly justifies the existence of a direct sales organization. So the problem is, how do these companies sell cloud services? Their existing channels are focused on $1M deals, so getting their attention on a $50,000 cloud services deal may prove to be problematic. Developing a new organization around selling cloud services may cause some major internal channel conflict.

These challenges may explain why there have been more announcements and declarations than actual cloud successes so far from the industry giants. That said, I would not bet against them, as cash is still king…but I suspect independent cloud divisions will have to be formed to achieve any quick success. It also means that lean and agile cloud start-ups that are evolving from the beginning around an on-demand model model have more of a fighting chance than start-ups in previous generations. I, for one, am looking forward to seeing how the cloud market plays out between the “Davids” and the “Goliaths.”

Consumer Websites Must Stop Cheating on Their Tests

Posted by Tom Lounibos on Nov 01, 2010. Leave a Comment

Another week goes by and more leading consumer websites crashed and burned, causing losses in both revenue (millions) and, perhaps more importantly, in consumer confidence in ecommerce. This week fail whales were sited swimming in the oceans of one of the world’s largest retailers, a leading children’s toy manufacturer and the industry’s leading online payment processor. PayPal alone (which was down for several hours this past Friday) may have lost hundreds of millions of dollars for itself and the enormous network of retail outlets that rely on them for their financial transaction services.

So why are fail whales continuing to happen so frequently? Don’t these companies test their sites? The answer, for load and performance testing, is of course they do…most of the time. In fact, some companies are spending millions of dollars on people, hardware and tools to load and performance test their websites. So why all the fail whales? One answer may be that organizations that have chosen the wrong testing tools or test service are, in fact, cheating on their tests!

Consumer facing website have become the primary channel of revenue and product information for millions of companies around the globe. So why cheat on testing? The pressure to deliver (business agility) is enormous today, for all IT organizations, and may well be a key reason cheating has become a common practice. Many test subcontractors and test companies cheat on their tests simply because they run out of time. According to PCWeek, this is what happened to AT&T on the pre-registration site for the iPhone4 launch. Due to a last minute feature upgrade they had no time to adequately performance test the site, which, of course, crashed an hour after it was launched, due to a 10x spike in traffic, creating a PR and revenue nightmare! Other organizations cheat because they just can’t afford the resources (people, hardware and tools) to properly test their sites in the first place.

The most common way to defend this cheating is to use semantics to obscure the shortcuts taken. When asked by an enraged business owner “did you test the site before going live?”, the answer is always “of course we did”. The problem is that it’s the wrong question. The right question is “did you test the site by accurately simulating real users performing both normal and unusual tasks, at and above expected volumes?”. For instance, if the goal was to simulate 5,000 concurrent users, a tester may respond that they tested for 5,000 “page views”! This is when language matters, since 5,000 page loads rarely equals the activity of 5,000 real users. In fact, it likely represents only a fraction of the target volume. By simply substituting page views or transactions for accurately simulating the activity of users on the site they almost certainly won’t reach the expected goal set by the business owner.

Another method of cheating the system is to adjust the timings of test scenarios. This practice is widely used by the testing community, primarily due to the cost of hardware and software when using traditional testing tools. For example, if buying a plane ticket online typically takes about 10 minutes, a clever tester may reduce the timing of this process in the test to just 1 minute. This, on the face of it, allows many more “users” into the system, but it doesn’t accurately simulate real world conditions. Finally, many leading edge companies are beginning to realize that the only way to accurately test a site is by including production testing. Testing only in the lab, and then extrapolating the results for the production environment, leaves far too many variables unaccounted for in the complex deployment environment that is the web. Again, cheating the testing system.

Whatever the reason for the cheating (lack of time, people, or resources) we must change this game now to maintain a high level of consumer confidence and continue to expand the growth of online commerce.

What Happens When Vendors Repackage Old Technology and Call it a Cloud Service?

Posted by Tom Lounibos on Oct 23, 2009. Leave a Comment

In an effort to remain relevant, some software vendors take marketing advantage of the newest hot technology fad at the expense of their own customers. Cloud computing (the newest hot trend) has definitely been defined and positioned by some traditional software and hardware vendor’s marketing organizations to meet their specific agenda, which usually means extending the life of an existing product or product line.  It has become a virtual “cloud rush” as to how many times “cloud computing” can be mentioned in product and marketing collateral. . . including collateral for products that were first developed back when Bill Clinton was president!

A great example of this “cloud rush” is Hewlett Packard with its LoadRunner product.  LoadRunner was developed in the early 90’s to help corporate development teams test client/server applications.  It became, over time, the de-facto standard testing tool for most enterprise companies and was priced accordingly.  Entry-level pricing began at $100,000 and if you needed to simulate thousands of users the cost skyrocketed into the millions of dollars very quickly.

Today, HP is attempting to “perfume the pig” so to speak, by repackaging LoadRunner into a new cloud-based service called Elastic Test. To the uneducated observer it simply looks like a new cloud service for testing web applications.  The problem is that it’s the same LoadRunner product built almost 20 years ago to test client/server applications and it carries a lot of technology baggage along with it. Subsequently, HP chooses to pass along a lot of this baggage in the form of costs back to its customer base.  For example, an entry-level HP virtual test will take weeks to develop and will have a “starting” cost of $45,000.  Hardly living up to cloud computing’s value proposition for on-demand services that provide ease, speed and affordability.

Newer players are quickly swooping in to fill in the gap left by HP’s lack of R&D in the cloud space. New players such as SOASTA, whose CloudTest service was built exclusively to leverage cloud computing for testing. It offers a low-cost, highly scalable and easy-to-use on-demand test model for customers.  As a comparison, the $45,000 “entry-level” HP Virtual Test that takes weeks to perform and deliver results, can be done within 24 hours at a cost of $750 using SOASTA’s On-Demand CloudTest service. This lower cost model also delivers greater quality testing. Traditionally, you may have only performed 6-8 performance tests on a client/server application because of the cost and time required.  Now, with true cloud testing services like SOASTA CloudTest, you can perform hundreds of tests for significantly less cost and in the same amount of time it takes to perform 6-8 traditional tests. This impact is significant, delivering greater reliability in your web applications.

Customers choosing to stay with HP as their test vendor out of loyalty will continue to have to deal with a 20 year old technology trying to conform to a 2009 eco-system, as well as pay a premium of up to 45X for that loyalty.  This, in a world where websites need more and more performance testing as they become more complex and we reach higher and higher spikes in user traffic.

Cloud computing is a “game changer” for customers seeking greater levels of web reliability. It enables new, leading-edge, agile and cost-effective cloud testing services.  However, be careful of impostors claiming to offer cloud services . . . more often then not you will be able to recognize them by their price tags and by the lack of quality in the actionable intelligence they deliver.  SOASTA CloudTest is the cloud leader in delivering the highest quality performance intelligence available on the market today.

NEW Web Services Performance Certification Program

Posted by Tom Lounibos on Jul 20, 2009. 1 Comment

I’m very excited to announce the new SOASTA Performance Certification Program designed to enable companies deploying software in the Cloud, at hosted data centers, or behind corporate firewalls to certify that their websites have been tested and have met or exceeded industry benchmarks for performance at peak levels of user traffic.  For the past ten years, the dirty little secret in the web development  community has been that whether due to cost, complexity or lack of resources, the vast majority of web applications and sites have not been tested at normal user volumes, much less for unexpected spikes in traffic. Which means our user communities have become the testers for virtually every website, a risk that has proven very costly time and again.

Times are changing–for every new service there are at least five competitors. Having a website that is slow to download or crashes frequently is no longer be tolerated.  “To our customers, performance matters!” said Lew Moorman, President, Rackspace Cloud. “By utilizing the SOASTA Performance Certification program our customers will have the ability to isolate performance issues before they occur by simulating real world user activity and traffic. Performance certification ensures that websites are being tested, which only leads to greater levels of reliability.”

Supporting SOASTA in this performance certification initiative are industry leaders in cloud computing including platform vendors, testing companies and independent cloud service providers. They include 3Tera, Appistry, Chegg.com, Enomaly, GoGrid, Hexaware Technologies, Intuit, JackBe, PowerTest, Rackspace, Rightscale, rPath and Zephyr. Concerns over performance and reliability are consistent themes heard by each of these companies when their customers face the uncertainties of extending access to applications outside the firewall, not to mention moving those applications completely offsite. This diverse group of companies represents all aspects of the application lifecycle, from development to deployment, all supporting the need for independent, external validation of end-to-end performance.

Many companies are currently considering or have already moved their applications to the Cloud, typically to lower costs and/or to take advantage of the elasticity of the compute power offered by the Cloud. However, just as in the data center, simply adding low cost servers doesn’t solve most performance problems. Performance-related issues, such as latency or an actual website crash, are often the result of a change to the deployment environment, changes in the application itself, or the inclusion of third-party content. The only way to certify a site’s performance is by simulating both expected and potential peak traffic while monitoring the impact. With Performance Certification, this is no longer too expensive or too complex to achieve and alleviates additional strain on internal resources.

“Improving performance cannot be achieved solely by adding hardware,” said Michael Crandell, CEO, Rightscale. “Web applications and deployment stacks are complex and must be tested under real world traffic conditions to assess what actual performance will be in production. Having a certification process that ensures sites are tested is an essential element of cloud computing development.”

We are very excited about this new program to certify websites and application performance! We are particularly appreciative of all of our partners who also feel that “Performance Matters” in delivering the best possible user experience.

SOASTA Performance Certification Program

Posted by Tom Lounibos on Jul 20, 2009. 2 Comments

SOASTA Performance Certification is designed as a turnkey process, minimizing disruption by eliminating any unnecessary impact on the existing environment. SOASTA and its partners collaborate with companies to define common use cases and then simulate those scenarios in the most accurate way possible—by using the web to test the web.

There are a number of purpose built tools to analyze the performance characteristics of individual components of a web-based application, such as web page design, application design, database implementation, or network architecture. SOASTA’s Performance Certification leverages its Global CloudTest™ Platform to provide an affordable end-to-end analysis of a site’s performance as well as measure responsiveness at normal and peak usage levels. Certified sites receive a comprehensive report on Key Performance Indicators (KPIs), with a focus on response times achieved at various user loads.

Certification confirms that the site has been tested at specific traffic volume levels (1K, 10K, 25K, 50K, or 100K users) and has been measured against KPIs such as response time and error rates. In addition, as part of the certification process SOASTA and its partners provide valuable analysis of site latency to help companies improve the overall performance and responsiveness of their website.

Companies such as Chegg.com, Dell, Intuit and Vovici are just a few of the companies that have had their websites SOASTA Certified. “As traffic to our website continues to grow at record levels, we need to ensure that our site will be able to meet the demand,” said Dan Bartow, Manager of Performance Engineering, Intuit TurboTax. “SOASTA enables us to simulate every possible user scenario including massive load. Thanks to SOASTA’s Performance Certification process, we are confident that our site will be ready for almost any situation we might encounter.”

Delivering Reliable Web Services Requires Web Scale Testing

Posted by Tom Lounibos on Apr 07, 2009. Leave a Comment

Who isn’t delivering a web service today is a much easier question to answer then asking who is. But the fact remains, if a web service is not reliable then it will not last long, no matter how interesting it might be. Ensuring a service’s reliability has been a goal of every developer, but it has been nearly impossible to attain until now.

The biggest obstacle has always been the cost of testing at web scale. Web scale is different for every company. In some cases, it’s a few hundred users hitting a web site at the very same time. For others, it’s millions of users (i.e. Facebook). No matter what your volume, it’s very expensive to achieve web scale in testing. For example, lets look at the environment you need to test 17,000 concurrent users of a web site. If you use some open source testing tools, you will most likely have to have over (200) server cores in your test lab. Try buying, configuring, and maintaining that many server cores and you will realize that “some assembly required” does not come close to describing the complexity and cost . . . and still you’re not done. You also need load balancers, monitors, dashboards, alerting systems, etc. . . . you get the picture! Even if you had unlimited money and manpower to achieve “web scale” in testing, you still have the problem of simulating traffic coming from inside your firewall and not externally as most Web2.0 applications function. For these reasons and many more, most companies have given up on testing their applications and network against the real world traffic that they are expecting (i.e. web scale).

That is, until today. Cloud testing enables companies to easily and affordably simulate “web scale” by using the access, availability, and affordability of Cloud Computing to create simulated web traffic. Companies like Hallmark, Intuit, Chegg and many many others are experiencing, for the first time, the ability test their web sites at “web scale” and beyond before their big day or event occurs. Cloud testing is changing how we test our web applications forever!

contact me at: tlounibos@soasta.com; twitter.com/lounibos

The Open Cloud Debate: A Customer’s Perspective

Posted by Tom Lounibos on Mar 27, 2009. Leave a Comment

My company SOASTA has provisioned over 30,000 cloud servers in the past year, using four different cloud platforms in doing so. Enterprise customers such as Intuit, Hallmark & Proctor & Gamble use our cloud testing service to load and performance test their web sites.

While I will not claim to be the largest cloud computing customer, I would assert that we know a little about the Cloud and its vendors. So, when I saw two recent blogs denouncing a new cloud initiative (one that is not even public yet) that calls for a movement toward an “Open Cloud” environment, I found their negative comments both amusing and disturbing. Amusing in that these individuals claiming to represent two leading edge cloud vendors were denouncing a new “idea” before it was even announced. Now that’s leading edge . . . and pretty amusing. I guess good ideas are only good when they are your ideas.

I also found their comments disturbing and lacking of in any real customer perspective on what we need from our cloud platform vendors in the first place. I became even more disturbed when one of the commentators was my leading cloud platform vendor . . . a vendor from whom we have provisioned nearly 27,000 servers with in the past year.

Dismissing an idea as not having merit without allowing it to see the light of day . . . being charitable . . . is a bit narrow. Why would these detractors shoot down openness in the first place? Seems a little like saying, “I’m against daylight”. You can complain all night, but the sun is still coming up in the morning.

As a CEO and a significant customer of cloud computing services, I would ask all the current and future platform vendors not to engage in traditional software politics and focus on what’s important. New ideas (especially in our industry and particularly in our current economy) should never be dismissed out of hand privately or publicly. They should be discussed and analyzed on their own merits, not dismissed on the basis of how they were introduced. I for one fully support this manifesto for an” Open Cloud Platform” and asked its authors to place my company’s name as one of its supporters. They have agreed to do so which suggests that the people who are pushing this open initiative are at least interested in how some “actual” customers of cloud computing may be thinking, while the companies who choose not to participate have sent a message to me (and others) that perhaps they are not.

contact me at: tlounibos@soasta.com; twitter.com/lounibos

Innovation Leads another Redistribution of Assets

Posted by Tom Lounibos on Mar 25, 2009. Leave a Comment

It appears that we are at the very early stage of what may become the largest redistribution of computing power since the late 80′s when we move from mainframe computers to desktops and servers. This triggered one of the biggest expansions in new jobs in the short history of the technology sector. Today, we once again face another market shift, and this time, the shift is away from the expensive and inefficient internal data centers model for a lower cost computing model called cloud computing. The cost of platforms for building, testing, and deploying web applications has become just too expensive to maintain internally and an alternative is required. Once again with new innovation, the opportunity to create a massive amount of new jobs is alive and well as we create new applications that will take advantage of this much lower cost delivery model.

Which brings me to my point, once again our technology sector is about to reinvent itself to meet a new market opportunity created by a business need and the emergence of technological innovation to meet this need.  In doing so, we will greatly expand our existing workforce to meet the potential of a much larger market opportunity.  Major shifts driven by innovation are not uncommon in the technology sector, why is this not true in other US industries? Wouldn’t it be nice to see in other sectors in our economy make similar market adjustments as quickly in the years to come. After all, innovation has always been one of the hallmarks of our country’s ability to lead. There are those who look at things the way they are, and ask why. We should dream of things that never were, and ask why not? (RFK).   It is time to embrace the innovation of cloud computing.

contact me at: tlounibos@soasta.com; twitter.com/lounibos

Email Us!
Subscribe to our Feed!
Join us on LinkedIn
Find us on Facebook
Follow our Tweets
See our pics